Performance of the World Health Organization (WHO) severe acute respiratory infection (SARI) case definitions in hospitalized children and youth: cross-sectional study

Peter J Gill; Caitlyn L Kaziev; Haifa Mtaweh; Tuana Kant; Claire Seaton; Daniel S Farrar; Hayley Wagman; Mei Han; Rohini R Datta; Sanjay Mahant; Gabrielle Freire; Aaron Campigotto; Jeffrey N Bone; Manish Sadarangani; Francine Buchanan; Shaun K Morris; READAPT-Kids Study

doi:10.1016/j.lana.2025.101034

. 2025 Feb 21;44:101034. doi: 10.1016/j.lana.2025.101034

Performance of the World Health Organization (WHO) severe acute respiratory infection (SARI) case definitions in hospitalized children and youth: cross-sectional study

Peter J Gill ^a,^b,^c,^d,^∗, Caitlyn L Kaziev ^b, Haifa Mtaweh ^b,^c,^e, Tuana Kant ^f, Claire Seaton ^g,^h,ⁱ, Daniel S Farrar ^b,^j, Hayley Wagman ^c, Mei Han ^k, Rohini R Datta ^j, Sanjay Mahant ^a,^b,^c,^d, Gabrielle Freire ^b,^c,^l, Aaron Campigotto ^m,ⁿ, Jeffrey N Bone ^o, Manish Sadarangani ^i,^p, Francine Buchanan ^b,^q, Shaun K Morris ^b,^c,^j,^r,^s; READAPT-Kids Study^t, for the

^aDivision of Paediatric Medicine, The Hospital for Sick Children, Toronto, Ontario, Canada

^bChild Health Evaluative Sciences, SickKids Research Institute, Toronto, Ontario, Canada

^cDepartment of Paediatrics, University of Toronto, Toronto, Ontario, Canada

^dInstitute of Health Policy, Management and Evaluation, University of Toronto, Toronto, Ontario, Canada

^eInterdepartmental Division of Paediatric Critical Care Medicine, The Hospital for Sick Children, Toronto, Ontario, Canada

^fTemerty Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada

^gDivision of Pediatric Hospital Medicine, BC Children’s Hospital, Vancouver, British Columbia, Canada

^hBC Children’s Hospital Research Institute, Vancouver, British Columbia, Canada

ⁱDepartment of Pediatrics, University of British Columbia, Vancouver, British Columbia, Canada

^jCentre for Global Child Health, The Hospital for Sick Children, Toronto, Ontario, Canada

^kClinical Research Unit, Children’s Hospital of Eastern Ontario Research Institute, Ottawa, Ontario, Canada

^lDivision of Pediatric Emergency Medicine, The Hospital for Sick Children, Toronto, Ontario, Canada

^mDivision of Microbiology, The Hospital for Sick Children, Toronto, Ontario, Canada

ⁿDepartment of Laboratory Medicine & Pathobiology, University of Toronto, Toronto, Ontario, Canada

^oBiostatistics Clinical Research Support Unit, BC Children’s Hospital Research Institute, Vancouver, British Columbia, Canada

^pVaccine Evaluation Center, BC Children’s Hospital Research Institute, Vancouver, British Columbia, Canada

^qCentre for Innovation and Excellence in Child and Family-Centred Care, The Hospital for Sick Children, Toronto, Ontario, Canada

^rDivision of Pediatric Infectious Diseases, The Hospital for Sick Children, Toronto, Ontario, Canada

^sDalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada

^∗

Corresponding author. Peter Gilgan Centre for Research and Learning, 686 Bay Street, Toronto, Ontario, M5G 0A4, Canada. peter.gill@sickkids.ca

Study group team members are listed in the Acknowledgements.

PMCID: PMC11904563 PMID: 40083965

Summary

Background

Respiratory tract infections with viral pathogens are frequently identified using the World Health Organization (WHO) case definition of severe acute respiratory infection (SARI), defined as fever of ≥38°Celsius, cough, onset within 10 days, and hospitalization. While there is extensive research in adults, less is known about the WHO SARI case definition performance in children and youth. We aimed to determine the performance of the WHO SARI and modified case definitions in identifying viral respiratory tract infections in hospitalized children and youth.

Methods

Retrospective observational cross-sectional study of hospitalized children (0–18 years) with an acute respiratory infection and who received a respiratory viral test at two large Canadian children’s hospitals from July 2022 to June 2023. The WHO SARI and modified SARI case definitions were evaluated overall, by virus and age, with reporting of sensitivity and specificity.

Findings

There were 2333 hospital admissions, with a median age of 2.4 years (IQR 0.8–5.0). 78% (n = 1828) had one or more viruses identified, most commonly respiratory syncytial virus (30%, n = 709). The WHO SARI definition had a sensitivity of 58% and specificity of 49% for identifying infections with a microbiologically confirmed virus. For Influenza only, the sensitivity was 71% and specificity 44%. The lowest sensitivity was among young children <3 months (28%) and 3 to <6 months (45%). Modified SARI definitions had similarly poor performance, with trade-offs of sensitivity and specificity.

Interpretation

The widely implemented WHO SARI case definition has sub-optimal performance among children and youth hospitalized with acute respiratory infections. Public health surveillance based on these case definitions may inadequately detect and monitor known and emerging infections, highlighting the need to develop an accurate and reliable SARI case definition for children and youth globally.

Funding

Public Health Agency of Canada, SickKids Foundation, BC Children’s Hospital.

Keywords: Severe acute respiratory infections, World Health Organization, Pandemics, Respiratory viruses, Children, Youth, Paediatrics, Hospitalization, Public health, Surveillance, Diagnostic accuracy, Sensitivity, Specificity, Case definitions

Research in context.

Evidence before this study

A literature search was conducted on MEDLINE, EMBASE, EBM Reviews, Elsevier SCOPUS and the WHO Global Index Medicus from database inception to June 7, 2024, for studies which reported sensitivity and specificity of any severe acute respiratory infections (SARI) case definitions in hospitalized paediatric patients. The search was conducted by combining search terms and subject headings, including “sensitivity and specificity”, “surveill∗”, “diagnos∗”, “Multiplex Polymerase Chain Reaction”, “Severe Acute Respiratory Infect∗”, “Respiratory Tract Infections”, “hospital admission∗”, “pediatric∗”, “child∗”, “youth∗”, “adolescent∗”. Of the 1080 articles identified, 46 were selected for full-text review, of which 12 studies were included. The 12 studies represented surveillance data from 63 sites across 8 countries using data from 2009 to 2016. The most common case definitions used were the 2014 WHO SARI (n = 8) and Integrated Management of Childhood Illness (IMCI) severe pneumonia (n = 6). Viral pathogens included influenza (n = 9) and RSV (n = 5). No studies evaluated SARS-CoV-2, and few studies evaluated other respiratory pathogens. The WHO SARI case definition had relatively low sensitivity in all cohorts, particularly among the youngest age groups. In studies that proposed modified SARI case definitions, sensitivity was only modestly improved.

Added value of this study

Our cross-sectional observational study leveraged detailed clinical data from 2333 hospitalizations of children and youth with acute respiratory infections at two large Canadian tertiary care paediatric hospitals from July 2022 to June 2023. To our knowledge, this study is the first to evaluate the performance of the WHO SARI case definition among children and youth using contemporary data, including during a time of high rates of hospital admissions, particularly among previously healthy children. Furthermore, this study is the first to evaluate the performance of several modified SARI case definitions, and the first to evaluate these case definitions with a broad panel of respiratory pathogens in children, including SARS-CoV-2. We found that the current WHO and modified SARI case definitions perform poorly in children, especially in detecting common childhood viruses, such as SARS-CoV-2. In particular, the case definitions had poor sensitivity among children <6 months, which are an important high-risk group for severe outcomes.

Implications of all the available evidence

Our study shows that the diagnostic test accuracy of current SARI surveillance definitions is sub-optimal in children and youth. These findings have important implications for global public health policy and emphasize that a more sensitive SARI case definition is needed for paediatrics. Without an accurate case definition, important populations of children (e.g., young infants), and settings where microbiologic surveillance may not be available (e.g., resource-limited settings), will not benefit from current surveillance and public health interventions, leading to inadequacies in care among the most vulnerable children and youth globally.

Introduction

Viral respiratory tract infections are a leading cause of hospitalization and death worldwide in children, most commonly due to respiratory syncytial virus (RSV) and influenza.1, 2, 3, 4 Sentinel surveillance systems for severe acute respiratory infections (SARI) play a crucial role in public health surveillance by monitoring trends and fluctuations in the trends in seasonal pathogens, and in assessing the impact of viral illnesses on healthcare utilization.5, 6, 7 Standardized case definitions help to reduce variability and facilitate the combination of data to account for under- and over-detection of a specific respiratory pathogen.⁸ SARI case definitions are especially useful in healthcare systems and settings where access respiratory viral testing for surveillance is limited, or before a pathogen-specific assay is developed.⁵^,⁷^,⁹ In 1999, the World Health Organization (WHO) first recommended a surveillance case definition for influenza-like illness (ILI) for those aged >5 years, consisting of sudden onset fever of ≥38 °C, and cough or sore throat.⁷ For children <5 years, the Integrated Management for Childhood Illness (IMCI) definition of ‘pneumonia’, originally developed for clinical care by community health workers in low- and middle-income countries, was suggested for surveillance.⁷ Due to sub-optimal performance of this definition in the 2009 H1N1 pandemic, in 2011, the WHO developed a new SARI surveillance definition, consisting of fever of ≥38 °C, cough, onset within the previous 7 days, and requiring hospitalization.⁵ As of 2014, the WHO SARI definition was modified to requiring symptom onset within the previous 10 days, and history of fever or measured fever of ≥38 °C, and has since become the accepted standard for detecting respiratory tract infections caused by viral pathogens.⁵ Of note, the WHO SARI case definition was developed for adults and there was no separate case definition for children.

Given that children have the highest incidence of respiratory tract infections,10, 11, 12 it is important that surveillance systems accurately detect and monitor cases in this population. However, the performance of the current WHO SARI case definition in children remains sub-optimal. A study of hospitalized children <2 years in Jordan from 2010 to 2013 evaluated the WHO SARI case definition and reported a sensitivity of 44% and specificity of 78% for detecting any respiratory virus, with improved sensitivity of 53% and reduced specificity of 61% for influenza.¹³ In Canada, the WHO SARI case definition had a sensitivity of 65% for influenza and 79% for other non-influenza respiratory viruses in hospitalized children <5 years old from 2012 to 2015.¹⁴ Several studies have reported that the accuracy of the WHO SARI case definition to detect RSV cases in hospitalized children is particularly poor, mainly due to the requirement of fever.¹⁵^,¹⁶ Given that the WHO SARI case definition was initially created for influenza surveillance, its performance may be inadequate for other viruses, such as RSV and SARS-CoV-2, in whom the infectious syndrome differs from influenza. As such, the WHO SARI case definition may lead to underreporting known pathogens or delayed detection of emerging respiratory illnesses in children who may be at higher risk for severe illness and who, given the high rates of infection in this age group, can be an important sentinel surveillance population.¹⁷^,¹⁸

Several authors have sought to improve the performance of the WHO SARI case definition with limited success.⁵^,¹³^,¹⁵^,17, 18, 19 For example, the WHO developed an “Extended SARI” case definition for RSV, which required cough or shortness of breath, and for those <6 months, included apnea and sepsis.²⁰ However, few studies have evaluated the WHO SARI case definition alongside modifications in one large cohort, and none have evaluated the performance in children with recent data, including with SARS-COV-2 infections. Contemporary accuracy data are critical to evaluate the utility of these case definitions with recent infectious agents given the large changes in epidemiology during and after the COVID-19 pandemic. Therefore, we sought to determine the performance of the WHO SARI and modified SARI case definitions in hospitalized children and youth with an acute respiratory infection.

Methods

Study design

We conducted a retrospective observational cross-sectional study of children (0–18 years) hospitalized with acute respiratory infections at two large Canadian tertiary care children’s hospitals, The Hospital for Sick Children (SickKids) in Toronto, Ontario and BC Children’s Hospital (BCCH) in Vancouver, British Columbia, from July 1, 2022 to June 30, 2023. Data were extracted from patient electronic medical records and managed in a Research Electronic Data Capture (REDCap) database.²¹ Research ethics board approval was obtained from both participating hospitals with a waiver of consent for secondary use of data. The study was reported according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline for observational studies.²²

Study population

Data were collected as part of the Clinical Characteristics and Outcomes of Hospitalized Children with Acute Respiratory Infections (READAPT-Kids) study, which aimed to describe the clinical outcomes and management of children aged <18 years admitted to hospital with acute respiratory infections. Eligible patients were initially identified using the most responsible discharge diagnostic International Classification of Diseases, Tenth Revision, Canada (ICD-10-CA) codes (Supplementary Table S1), derived from a modified version of the Paediatric Clinical Classification System (PECCS).²³ Patient charts were then manually screened for inclusion by trained research assistants. Training for study staff included a detailed study overview, training on electronic health record navigation, protocols for data collection with standardized definitions, hands-on practice on test cases, random chart audits, and regular feedback to ensure accuracy and consistency.

Patients were included if they were aged <18 years at the time of hospitalization, were admitted for an acute respiratory illness, and had a suspected or microbiologically confirmed viral or bacterial pathogen. Acute respiratory illnesses were broadly defined as bronchiolitis, croup, asthma exacerbation, pneumonia (viral or bacterial), including complicated pneumonia, acute respiratory distress syndrome, or respiratory failure. Patients were excluded if the hospital admission was elective, post-operative, or due to trauma, or if patients had a hospital acquired infection, as defined by local infection prevention and control criteria. Patients with repeat hospital admissions more than 30 days apart were eligible for inclusion as distinct cases.

Given that respiratory viral testing would confirm infections, only children who received a microbiologic respiratory test (either a multiplex polymerase chain reaction [PCR] or rapid antigen test [RAT]) were included in this study, and only those with complete datasets (Supplementary Fig. S1).

Data collection

Detailed demographic and clinical data were collected, including symptoms at initial presentation. Fever was captured subjectively by parent report of tactile fever or measured fever ≥38.0 °C at home. Respiratory viral testing results were collected, which included RAT and/or multiplex PCR tests. RAT involved reporting of RSV, SARS-CoV-2, and influenza (A and B), while multiplex PCR tests involved reporting these and enterovirus/rhinovirus (EV/RV), parainfluenza (PIV) (1, 2, 3, 4), human metapneumovirus (hMPV), adenovirus (AdV), bocavirus (BoV), and seasonal coronaviruses (sCoV) (229E, NL63, OC43, HKU1). Both RAT and multiplex PCR tests were considered for detection of RSV, SARS-CoV-2, and influenza, while only multiplex PCR tests were considered for the other viruses. Enterovirus and rhinovirus, as well as seasonal coronavirus strands OC43 and HKU1, could not be distinguished by testing at one site, and thus were grouped together, respectively. At both sites, viral testing was completed at the discretion of the clinical care provider (e.g., no hospital wide policy to test all hospitalized patients), and most children were tested with a multiplex PCR, while a RAT was commonly performed at outside community hospitals. For transferred patients, viral testing was repeated at the discretion of the clinical care provider.

Severe acute respiratory infection (SARI) case definitions

The WHO SARI case definition is defined as 1) reported or measured fever of ≥38° Celsius; 2) cough; 3) symptom onset within the previous 10 days; and 4) requiring hospitalization.⁵ We also evaluated published modified SARI case definitions, which included at least one of the following: fever (reported or measured ≥38 °C), cough, sore throat, increased work of breathing (WOB), and shortness of breath (SOB).⁵^,¹³^,¹⁵^,17, 18, 19, 20 These modified SARI case definitions included the WHO “Extended SARI” case definition for hospital-based surveillance for severe RSV infection.²⁰ Full descriptions are outlined in Supplementary Table S2.

Data analysis

Descriptive statistics were used to summarize patient characteristics, including medians and interquartile ranges (IQR) for continuous variables, and frequencies and percentages for categorical variables.

To assess the performance of each SARI case definition in predicting test-positive viral infections, we tabulated the number of patients meeting versus not meeting the SARI criteria against the number of patients testing positive versus testing negative for viral pathogens. Analyses for SARS-CoV-2, influenza, and RSV were restricted to those who had any pathogen-specific testing (multiplex PCR or RAT). Analyses for any other viral pathogen were restricted to those patients who had multiplex PCR testing conducted. Where a patient had both a respiratory multiplex PCR and RAT, the multiplex PCR test was taken as the “gold standard” due to its higher sensitivity and specificity.²⁴ Any indeterminate test results were considered as negative. For any viral co-infections, patients were considered positive for any identified viruses. Sensitivity analyses were conducted for mono-infections to evaluate virus-specific performance of SARI case definitions.

Measures of performance were calculated for each combination of SARI case definition and any respiratory viral pathogen, including sensitivity and specificity. This was repeated for each viral respiratory pathogen, and for each age subgroup (i.e., <3 months, 3–<6 months, 6–<12 months, 12–<24 months, 2–<5 years, and 5–<12 years, 12–<18 years).

We also evaluated the performance of each included SARI case definition using logistic regression and a receiver operating characteristic (ROC) curve analysis. The logistic regression model adjusted for sex, and age as a categorical variable. Diagnostic accuracy statistics were reported as above.²⁵ Thresholds for diagnostic accuracy statistics were determined using the Youden index to maximize sensitivity and specificity. All data analysis was conducted in R version 4.3.1 without associated 95% confidence intervals as we evaluated the case definitions in the full study population.

Role of the funding source

The funders had no role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.

Results

Description of study population

There were 2333 eligible patients, of which 60% were male (n = 1396) and with a median age of 2.4 years (IQR 0.8–5.0 years) (Table 1, Supplementary Table S3). Half of patients had a pre-existing chronic condition, with the most common being gastrointestinal disease (14%, n = 333), neurological disorders (14%, n = 332), and genetic disease (14%, n = 331), with a detailed breakdown by age in Supplementary Table S4. A viral pathogen was detected in 78% of patients (n = 1828), with RSV being the most common virus (30%, n = 709), followed by enterovirus/rhinovirus (26%, n = 599). Approximately one-quarter (23%; n = 535) had a viral co-infection.

Table 1.

Demographic characteristics and viral etiology of study population by WHO SARI status.

Characteristic	All patients, N = 2333	WHO SARI-positive, N = 1359	WHO SARI-negative, N = 974
Age (years), median IQR	2.4 (0.8, 5.1)	2.6 (1.2, 4.8)	1.9 (0.3, 5.6)
Age group, n (%)
<1 months	111 (4.8%)	25 (1.8%)	86 (8.8%)
1–<3 months	223 (9.6%)	78 (5.7%)	145 (15%)
3–<6 months	123 (5.3%)	60 (4.4%)	63 (6.5%)
6–<12 months	192 (8.2%)	120 (8.8%)	72 (7.4%)
12–<24 months	376 (16%)	249 (18%)	127 (13%)
2–<5 years	712 (31%)	501 (37%)	211 (22%)
5–<12 years	481 (21%)	275 (20%)	206 (21%)
12–<15 years	69 (3.0%)	30 (2.2%)	39 (4.0%)
15–<18 years	46 (2.0%)	21 (1.5%)	25 (2.6%)
Sex, n (%)
Female	937 (40%)	540 (40%)	397 (41%)
Male	1396 (60%)	819 (60%)	577 (59%)
Chronic comorbid condition present^a, n (%)	1177 (50%)	661 (49%)	516 (53%)
Neurological disorders	332 (14%)	166 (12%)	166 (17%)
Gastrointestinal disease	333 (14%)	183 (13%)	150 (15%)
Genetic disease (e.g., Trisomy 21)	331 (14%)	183 (13%)	148 (15%)
Asthma	321 (14%)	171 (13%)	150 (15%)
Developmental disorder (e.g., cerebral palsy)	310 (13%)	173 (13%)	137 (14%)
Cardiac disease, including congenital heart disease	292 (13%)	171 (13%)	121 (12%)
Pulmonary disease (not asthma)	291 (12%)	144 (11%)	147 (15%)
Complex care program	259 (11%)	131 (9.6%)	128 (13%)
Others^b	174 (7.5%)	109 (8.0%)	65 (6.7%)
SARS-CoV-2 testing^c	2293 (98%)	1335 (98%)	958 (98%)
RSV testing^c	2053 (88%)	1177 (87%)	876 (90%)
Influenza testing^c	2056 (88%)	1179 (87%)	877 (90%)
Respiratory panel testing^d	1930 (83%)	1099 (81%)	831 (85%)
Any viral pathogen identified^c, n (%)	1828 (78%)	1062 (78%)	766 (79%)
SARS-CoV-2	126 (5.4%)	65 (4.8%)	61 (6.3%)
Influenza (Types A and B)	168 (7.2%)	120 (8.8%)	48 (4.9%)
Respiratory syncytial virus	709 (30%)	426 (31%)	283 (29%)
Parainfluenza (Types 1, 2, 3, and 4)	171 (7.3%)	107 (7.9%)	64 (6.6%)
Enterovirus/Rhinovirus	599 (26%)	289 (21%)	310 (32%)
Human metapneumovirus	195 (8.4%)	147 (11%)	48 (4.9%)
Bocavirus	45 (1.9%)	32 (2.4%)	13 (1.3%)
Adenovirus	100 (4.3%)	61 (4.5%)	39 (4.0%)
Seasonal coronavirus (229E, NL63, and OC43/HKU1)	101 (4.3%)	62 (4.6%)	39 (4.0%)
Number of viral pathogens identified
Zero (negative respiratory viral testing)	505 (22%)	297 (22%)	208 (21%)
One	1290 (55%)	775 (57%)	515 (53%)
Two	403 (17%)	206 (15%)	197 (20%)
≥Three	135 (5.8%)	81 (6.0%)	54 (5.5%)

Open in a new tab

WHO = World Health Organization; SARI = Severe acute respiratory infection; IQR = Interquartile range; RSV = Respiratory syncytial virus.

Indicates the most prevalent chronic conditions.

Other chronic conditions include enrollment in a complex care program, use of puffers/inhalers at home, sickle cell anemia, immunodeficiency, use of immunosuppressive medication, use of biologic drugs, morbid obesity, malignancy, use of home oxygen, chronic kidney disease, haematological disease, type 1 diabetes, rheumatological disorder, inborn error of metabolism, and others not listed.

Denotes the use of multiplex polymerase chain reaction (PCR) test and/or rapid antigen test (RAT) performed.

Denotes the use of multiplex PCR testing only.

Clinical features

Cough was reported in most patients (83%, n = 1933), followed by fever (measured or reported ≥ 38 °C) (70%, n = 1635), and increased work of breathing (69%, n = 1604) (Table 2). Most patients (94%, n = 2199) presented with other symptoms, with coryza/nasal congestion (61%, n = 1433) being the most common. Shortness of breath (18%, n = 422) and sore throat (3%, n = 70) were less common (Supplementary Table S5). A description of clinical features by age sub-groups is outlined in Supplementary Tables S6–S12.

Table 2.

Description of clinical signs and SARI symptoms for all patients by virus type.

Characteristic	All patients, N = 2333	RSV, N = 709	SARS-CoV-2, N = 126	Influenza, N = 168	PIV, N = 171	EV/RV, N = 599	hMPV, N = 195	AdV, N = 100	BoV, N = 45	sCOV, N = 101	Negative, N = 505
Time since symptom onset (days), median (IQR)	4 (2, 6)	4 (3, 5)	3 (2, 4)	5 (3, 7)	4 (2, 6)	3 (2, 5)	4 (3, 6)	5 (3, 7)	3 (2, 5)	3 (2, 7)	4 (2, 6)
Number of symptoms at presentation, median (IQR)	5 (4, 6)	5 (4, 6)	5 (3, 6)	5 (4, 7)	5 (4, 6)	5 (4, 6)	5 (4, 7)	5 (4, 7)	5 (4, 6)	5 (4, 6)	5 (4, 7)
SARI symptoms at presentation, n (%)	2314 (99%)	707 (100%)	124 (98%)	166 (99%)	170 (99%)	590 (98%)	193 (99%)	100 (100%)	44 (98%)	101 (100%)	502 (99%)
Cough	1933 (83%)	629 (89%)	94 (75%)	146 (87%)	137 (80%)	473 (79%)	172 (88%)	80 (80%)	42 (93%)	79 (78%)	402 (80%)
Fever (any)	1635 (70%)	482 (68%)	90 (71%)	138 (82%)	132 (77%)	363 (61%)	166 (85%)	76 (76%)	34 (76%)	81 (80%)	369 (73%)
Fever (measured ≥ 38 °C)	657 (28%)	187 (26%)	38 (30%)	53 (32%)	55 (32%)	155 (26%)	69 (35%)	35 (35%)	11 (24%)	34 (34%)	139 (28%)
Fever (reported)	1156 (50%)	347 (49%)	60 (48%)	99 (59%)	91 (53%)	241 (40%)	114 (58%)	49 (49%)	25 (56%)	53 (52%)	280 (55%)
Increased work of breathing	1604 (69%)	536 (76%)	87 (69%)	101 (60%)	122 (71%)	424 (71%)	117 (60%)	67 (67%)	38 (84%)	59 (58%)	327 (65%)
Shortness of breath	422 (18%)	80 (11%)	20 (16%)	37 (22%)	16 (9.4%)	118 (20%)	46 (24%)	23 (23%)	4 (8.9%)	8 (7.9%)	109 (22%)
Sore throat	70 (3.0%)	8 (1.1%)	2 (1.6%)	8 (4.8%)	5 (2.9%)	22 (3.7%)	3 (1.5%)	3 (3.0%)	0 (0%)	1 (1.0%)	24 (4.8%)
Symptom onset within 10 days, n (%)	2194 (94%)	683 (96%)	123 (98%)	143 (85%)	161 (94%)	563 (94%)	188 (96%)	91 (91%)	41 (91%)	96 (95%)	466 (92%)
Meets WHO SARI case definition^a, n (%)	1359 (58%)	426 (60%)	65 (52%)	120 (71%)	107 (63%)	289 (48%)	147 (75%)	61 (61%)	32 (71%)	62 (61%)	297 (59%)

Open in a new tab

RSV = Respiratory syncytial virus; PIV = Parainfluenza virus; EV/RV = Enterovirus/rhinovirus; hMPV = Human metapneumovirus; AdV = Adenovirus; BoV = Bocavirus; sCOV = Seasonal coronaviruses; IQR = Interquartile range; WHO = World Health Organization; SARI = Severe acute respiratory infection.

All patients were considered as having symptoms within 10 days of hospital admission, given that each case was an acute presentation of respiratory infection.

WHO SARI case definition

The WHO SARI definition was met in 58% (n = 1359) of patients, with a sensitivity of 58%, and specificity of 49% (Table 3). The performance of the WHO SARI definition varied considerably by age, with the highest sensitivity among children aged 2–5 years (70%) and lowest in those <3 months (28%) (Table 4). The specificity was highest in children <3 months (81%) and lowest in children aged 12–24 months (33%). Sensitivity analysis excluding patients with a viral co-infection reported similar results overall and by age sub-groups (Supplementary Tables S13 and S14).

Table 3.

Diagnostic performance of the WHO SARI case definition by virus type for all ages.

Respiratory Virus	Sensitivity	Specificity
Any respiratory virus^a	0.58	0.49
Any one of RSV, SARS-CoV-2, or Influenza^b	0.56	0.39
Any one of RSV or influenza^b	0.54	0.38
SARS-CoV-2	0.52	0.41
RSV	0.60	0.44
Influenza	0.71	0.44
Any one of the other viruses^a	0.59	0.46
Parainfluenza	0.63	0.44
Enterovirus/rhinovirus	0.48	0.39
Human metapneumovirus	0.75	0.45
Adenovirus	0.61	0.43
Bocavirus	0.71	0.43
Seasonal coronavirus	0.61	0.43

Open in a new tab

WHO = World Health Organization; SARI = Severe acute respiratory infection; RSV = Respiratory syncytial virus.

Respiratory multiplex polymerase chain reaction (PCR) test only performed.

Respiratory multiplex PCR test and/or rapid antigen test (RAT) performed.

Table 4.

Diagnostic performance of the WHO SARI case definition for any respiratory virus overall and by age group from multiplex PCR testing.

Age group	Sensitivity	Specificity
All ages (0–18 years)	0.58	0.49
<3 months	0.28	0.81
3–<6 months	0.45	0.53
6–<12 months	0.62	0.59
12–<24 months	0.66	0.33
2–<5 years	0.70	0.35
5–<12 years	0.58	0.51
12–<18 years	0.53	0.66

Open in a new tab

WHO = World Health Organization; SARI = Severe acute respiratory infection; PCR = Polymerase chain reaction.

WHO SARI case definition by age and virus

In evaluating the WHO SARI definition by virus, the highest sensitivity was in children and youth with human metapneumovirus (75%), with the lowest sensitivity among children with enterovirus/rhinovirus (48%) (Supplementary Fig. S2). The highest sensitivity by virus and age was 92% for influenza among children aged 6–12 months (Supplementary Table S15). Specificity was consistently poor across all ages and viruses, with the lowest at 24% for children 2–5 years with enterovirus/rhinovirus.

Modified SARI case definitions

There was wide variability in performance of the modified SARI case definitions overall (Fig. 1, Supplementary Table S16) and by age group (Supplementary Tables S17 and S18). Having either cough or increased work of breathing,¹⁵^,¹⁹ had the highest sensitivity of 95%, but the lowest specificity at 12%. The WHO “Extended SARI” case definition for hospital-based surveillance for severe RSV infection had high sensitivity (84% for cough only, and 86% for cough or shortness of breath) but low specificity. Having measured fever ≥ 38 °C and cough,¹⁷^,¹⁹ had the highest specificity (75%), but the lowest sensitivity at 24%. In general, sensitivity was inversely proportional to specificity across the modified SARI case definitions. The performance of the WHO SARI and modified case definitions improved modestly when evaluated using multivariable logistic regression (Supplementary Table S19).

Discussion

We evaluated the performance of the WHO SARI case definition and published modified SARI definitions in a large cross-sectional study of hospitalized children with acute respiratory infections. The WHO SARI case definition had poor performance overall, particularly in young children <3 months of age. Similarly, modified SARI definitions did not show improvements in performance compared to the WHO SARI case definition. The main purpose of a SARI case definition is to assist with surveillance to understand trends in disease severity, changes in disease epidemiology and disease burden of known infections, and for WHO SARI specifically, to evaluate the impact of influenza in relation to other diseases.⁵ An accurate SARI case definition is particularly important in settings where microbiologic testing for surveillance is limited (e.g., due to limited testing availability or testing that is not yet available for an emerging virus), and to aid in sentinel surveillance, which is critical for effective pandemic preparedness and response.⁵^,⁷ Our findings highlight the importance of understanding the contemporary performance of a SARI case definition in several viral pathogens in paediatric patients, who are often severely impacted by acute respiratory infections.

The performance of the WHO SARI definition in our study differs from prior paediatric studies likely due to important differences in age groups, viruses, settings, and time-periods. Amini et al. evaluated the WHO SARI definition for children <5 years with influenza in Quebec between 2012 and 2015 and reported a sensitivity of 65.3%,¹⁴ compared to our sensitivity of 76%. While Makokha et al. reported superior performance of WHO SARI for children <5 hospitalized with influenza in Kenya from 2009 to 2013 with a sensitivity of 84%,⁹ Rha et al. reported similar performance as our study for children <5 years with RSV in South Africa from 2009 to 2014.¹⁵ A large proportion of our patients in our study also had underlying conditions, compared to the relatively healthy cohorts reported by others in different settings. The heterogeneity of our population could account for poorer performance of the WHO SARI case definition, given differences in presentation of symptoms, need to seek care, or thresholds for admission to hospital, since having multiple underlying conditions is associated with increased severity of illness compared to healthy children, particularly for RSV.26, 27, 28, 29

In our paediatric population, the sensitivity of the WHO SARI case definition was lower than in adult studies. This can be attributed to differences in symptoms, virus prevalence, admission thresholds, settings, and time-periods. In patients aged 18–64 years, Davis et al. reported a sensitivity of 86% for adults with influenza and a sensitivity of 79% for RSV in New Zealand between 2012–2016,¹⁸ compared to our cohort sensitivity of 72% for influenza and 60% for RSV. However, Rowlinson et al. reported that 78% of adults ≥ 65 years hospitalized with influenza in Egypt between 2009 and 2012 met the WHO SARI case definition,⁶ similar to our study for children with influenza. The differences in performance between paediatric and adult cohorts is likely attributable to differences in symptoms at presentation.³⁰ In particular, fever, which is a core feature of the WHO SARI case definition, is less commonly reported in young children. The requirement of fever in the WHO SARI and modified SARI definitions improves specificity but lowers sensitivity, especially in young children, and in certain infections, such as RSV and enterovirus/rhinovirus.³¹ While the performance of the WHO SARI case definition differs between adults and children, most contemporary adult studies focus only on influenza and RSV. Although these are common in children, several other viruses, such as enterovirus and rhinovirus, are highly prevalent in hospitalized children, yet are less well studied, despite reporting the worst performance by the WHO SARI case definition in our study.

There were large differences in diagnostic performance across modified SARI definitions, including across age groups. In our study, the modified SARI case definitions that did not include fever – specifically, had cough only (84%),¹³^,¹⁵^,¹⁸^,¹⁹ cough or shortness of breath (86%),¹⁸ or cough or increased work of breathing (95%),¹⁵^,¹⁹ – had the highest sensitivity. Sensitivity increases when fever is removed; however, it also decreases specificity. This is an accurate description of our population, given that cough (n = 1,933, 83%) and increased work of breathing (n = 1,604, 69%) were among the most common symptoms in children hospitalized with acute respiratory infections. Our findings are similar to others who reported that cough had the strongest association with respiratory infection compared to fever, suggesting that requiring fever or cough has a major impact on diagnostic performance.¹⁵^,³¹

Based on the poor diagnostic performance of the WHO SARI case definition, particularly in detecting common pathogens in children, a large proportion of children with acute respiratory infections requiring hospitalization are not detected using the current definition. Further, given that the poorest performance was among one of the highest risk groups – children <3 months – it is imperative that a more sensitive case definition be developed for paediatrics. High sensitivity is particularly important to avoid underestimating disease burden and allow for adequate policy planning and healthcare utilization for future respiratory viral seasons.¹⁶ This is equally critical in resource-limited regions and health systems, where SARI case definitions are relied heavily upon to detect and track respiratory illnesses in children.⁵^,⁷^,⁹ Future studies should look at specific variations in signs and symptoms that may produce a more sensitive SARI definition across multiple viruses, specifically in paediatrics, and could benefit from exploring symptoms that are most prevalent in common childhood viruses, including enterovirus/rhinovirus and RSV.³⁰ Furthermore, such studies may benefit from predictive modelling that considers both positive and negative predictors, which would help in both detecting and ruling out viral pathogens. While predictive modelling may not be feasible in limited resource settings, with the growing use of electronic medical records, it may become more efficient and cost-effective in the future to automate detection. These models could be used to detect infections, and predict illness and severity, which has important implications for public health, pandemic preparedness, and in detecting and predicting respiratory illness trends in children.

This study has several strengths, including a large cohort with detailed clinical information from several paediatric age groups, and is one of the first to evaluate the performance of multiple SARI case definitions with a broad panel of respiratory pathogens, including SARS-CoV-2. While our study was designed to compare the performance of the SARI case definition for distinguishing viral from non-viral infections, the SARI case definition may be used to monitor other trends, such as in atypical infections like Mycoplasma pneumoniae.³² The study includes children from two large tertiary care hospitals reflecting varying epidemiology in different regions and populations. Each is a large urban metropolis city with a high population density, and diverse multi-cultural population with 40–50% of its population born outside of Canada. However, there are important limitations, including that the population may not be representative of other regions in Canada, particularly those that are rural and remote, with lower population density, and with smaller immigrant populations. While the included hospitals function as the primary hospital for a large population of children and youth who live in the city, 40% of the patients were transferred from other hospitals. Further, the study was conducted during a surge of acute respiratory infections, which may limit generalizability to other time periods. For some patients that were transferred, viral testing results were based on outside hospitals, and testing was not always repeated, unless deemed clinically necessary. However, this is often representative of variations in clinical practices, and further emphasizes the need for a reliable SARI case definition that can be used to detect respiratory viral infections in children and youth across different settings and systems globally. Data regarding ethnicity were also not available. Lastly, 50% of the included patients had a reported chronic condition, which is higher than 30% reported in the Canadian population of children and youth, which may impact generalizability.

Conclusion

In our multi-site cross-sectional study of over 2300 children and youth hospitalized with an acute respiratory infection, the WHO SARI case definition had a modest sensitivity of 58% and specificity of 49%. Given that the performance of the WHO SARI and modified SARI case definitions is sub-optimal in paediatrics, these findings emphasize the need to develop a SARI case definition that is more accurate and reliable. Developing such a surveillance tool can be used to better detect emerging viral cases, which in turn, can better manage healthcare utilization and facilitate timely response to emerging infections globally.

Contributors

All authors of this study participated in the concept, study design, and interpretation of data. Dr. Gill, Ms. Kaziev, Ms. Han, Ms. Kant, Mr. Farrar, and Dr. Morris contributed to the acquisition and/or analysis of the data. Dr. Gill, Ms. Han, and Ms. Kant had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. The manuscript was drafted by Dr. Gill and Ms. Kaziev, and then was critically revised for important intellectual content by all authors of this study.

Data sharing statement

Deidentified patient data are available from the authors upon request. The full dataset creation plan is available from the authors upon request.

Declaration of interests

Dr Gill has received grants from the Canadian Institute of Health Research (CIHR), the PSI Foundation, and The Hospital for Sick Children; he has received nonfinancial support from the CIHR Institute of Human Development, Child and Youth Health (as a member of the institute advisory board, expenses reimbursed to attend meetings); and he is a member of the Hospital Pediatrics and BMJ Evidence Based Medicine Editorial Board. Dr Mahant has received grants from the CIHR, the PSI Foundation, and The Hospital for Sick Children, and is a member of the Journal of Hospital Medicine Editorial Board. Dr Morris reports speaker fees from GlaxoSmithKline Canada and Sanofi-Pasteur, and has served on ad hoc advisory boards for GlaxoSmithKline Canada, Pfizer and Sanofi-Pasteur, all unrelated to this study. Dr Sadarangani is supported via salary awards from the BC Children’s Hospital Foundation and Michael Smith Health Research BC, and has been an investigator on projects funded by GlaxoSmithKline, Merck, Moderna, Pfizer and Sanofi-Pasteur; all funds have been paid to his institute, and he has not received any personal payments. Dr. Buchanan has received grants from the CIHR, support from the Academy of Medical Sciences (UK) to attend meeting, and honoraria from the American Academy for Cerebral Palsy and Developmental Medicine to attend conference. Dr. Seaton has received support from the Paediatric Inpatient Research Network to attend meeting. All other authors declare no conflict of interest.

Acknowledgements

This study was funded by the Public Health Agency of Canada, SickKids Foundation, and the BC Children's Hospital. We would also like to thank the extensive READAPT-Kids Study team for their contributions.

Additional non-author contributors of the READAPT-Kids Study include:

•
Children’s Hospital of Eastern Ontario (CHEO), Ottawa, Ontario, Canada: Nicholas Barrowman.
•
The Hospital for Sick Children (SickKids), Toronto, Ontario, Canada: Rae S. M. Yeung, Anya Nair, Nafisa Anwar, Rizk ElMadbak, Nardin Kirolos, Jonathan Fortini, Kody M. Wolfstadt, Nilushi de Silva, Polina Kyrychenko, Shamama Raza, Vincent Flores, Keane Fuerte, Pardis Noormohammadpour, Hafsa Azher.
•
BC Children’s Hospital (BCCH), Vancouver, British Columbia, Canada: Jennifer Retallack, Jocelyn A. Srigley, Thomas McLaughlin, Candice Wiedman, Melissa Braschel, Alam Lakhani, Opninder Lindstrom, Sanja Hadzi-Nikolova, Min Jung Kim, Victoria Tapics, Henry Okpaladigbo, Joanna Xu, Zainab Zeyan, Baneesh Khosa.

Footnotes

^{Appendix A}

Supplementary data related to this article can be found at https://doi.org/10.1016/j.lana.2025.101034.

Contributor Information

Peter J. Gill, Email: peter.gill@sickkids.ca.

READAPT-Kids Study:

Nicholas Barrowman, Rae S.M. Yeung, Anya Nair, Nafisa Anwar, Rizk ElMadbak, Nardin Kirolos, Jonathan Fortini, Kody M. Wolfstadt, Nilushi de Silva, Polina Kyrychenko, Shamama Raza, Vincent Flores, Keane Fuerte, Pardis Noormohammadpour, Hafsa Azher, Jennifer Retallack, Jocelyn A. Srigley, Thomas McLaughlin, Candice Wiedman, Melissa Braschel, Alam Lakhani, Opninder Lindstrom, Sanja Hadzi-Nikolova, Min Jung Kim, Victoria Tapics, Henry Okpaladigbo, Joanna Xu, Zainab Zeyan, and Baneesh Khosa

Appendix ASupplementary data

Supplementary Figs. S1 and S2 and Tables S1–S19

mmc1.docx^{(508.2KB, docx)}

References

1.Kyu H.H., Vongpradith A., Sirota S.B., et al. Age–sex differences in the global burden of lower respiratory infections and risk factors, 1990–2019: results from the Global Burden of Disease Study 2019. Lancet Infect Dis. 2022;22(11):1626–1647. doi: 10.1016/S1473-3099(22)00510-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Brady M., Duffy R., Domegan L., et al. Establishing severe acute respiratory infection (SARI) surveillance in a sentinel hospital, Ireland, 2021 to 2022. Euro Surveill. 2023;28(23) doi: 10.2807/1560-7917.es.2023.28.23.2200740. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Perin J., Mulick A., Yeung D., et al. Global, regional, and national causes of under-5 mortality in 2000–19: an updated systematic analysis with implications for the sustainable development goals. Lancet Child Adolesc Health. 2022;6(2):10–115. doi: 10.1016/S2352-4642(21)00311-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Williams B.G., Gouws E., Boschi-Pinto C., Bryce J., Dye C. Estimates of world-wide distribution of child deaths from acute respiratory infections. Lancet Infect Dis. 2002;2(1):25–32. doi: 10.1016/S1473-3099(01)00170-0. [DOI] [PubMed] [Google Scholar]
5.Fitzner J., Qasmieh S., Mounts A.W., et al. Revision of clinical case definitions: influenza-like illness and severe acute respiratory infection. Bull World Health Organ. 2018;96(2):122–128. doi: 10.2471/BLT.17.194514. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Rowlinson E., Peters L., Mansour A., et al. Comparison of common acute respiratory infection case definitions for identification of hospitalized influenza cases at a population-based surveillance site in Egypt. PLoS One. 2021;16(3) doi: 10.1371/journal.pone.0248563. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Institute of Medicine (US) Forum on Microbial Threats . National Academies Press; 2007. Global infectious disease surveillance and detection. [DOI] [PubMed] [Google Scholar]
8.Khamphaphongphane B., Chiew M., Mott J.A., et al. Estimating the national burden of hospitalizations for influenza-associated severe acute respiratory infection in the Lao People’s Democratic Republic, 2016. West Pac Surveill Response J. 2021;12(2) doi: 10.5365/wpsar.2020.11.2.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Makokha C., Mott J., Njuguna H.N., et al. Comparison of severe acute respiratory illness (sari) and clinical pneumonia case definitions for the detection of influenza virus infections among hospitalized patients, western Kenya, 2009-2013. Influenza Other Respir Viruses. 2016;10(4):333–339. doi: 10.1111/irv.12382. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Galanti M., Birger R., Ud-Dean M., et al. Longitudinal active sampling for respiratory viral infections across age groups. Influenza Other Respir Viruses. 2019;13(3):226–232. doi: 10.1111/irv.12629. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Tsagarakis N.J., Sideri A., Makridis P., Triantafyllou A., Stamoulakatou A., Papadogeorgaki E. Age-related prevalence of common upper respiratory pathogens, based on the application of the Film Array Respiratory panel in a tertiary hospital in Greece. Medicine. 2018;97(22) doi: 10.1097/MD.0000000000010903. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Mandelia Y., Procop G.W., Richter S.S., Worley S., Liu W., Esper F. Dynamics and predisposition of respiratory viral co-infections in children and adults. Clin Microbiol Infection. 2021;27(4):631.e1–631.e6. doi: 10.1016/j.cmi.2020.05.042. [DOI] [PubMed] [Google Scholar]
13.Klink T., Rankin D.A., Piya B., et al. Evaluating the diagnostic accuracy of the WHO severe acute respiratory infection (SARI) criteria in Middle Eastern children under two years over three respiratory seasons. PLoS One. 2020;15(4) doi: 10.1371/journal.pone.0232188. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Amini R., Gilca R., Douville-Fradet M., Boulianne N., De Serres G. Evaluation of the new world health organization case definition of severe acute respiratory infection for influenza surveillance during the peak weeks of two influenza seasons in Quebec, Canada. J Pediatric Infect Dis Soc. 2017;6(3):297–300. doi: 10.1093/jpids/piw044. [DOI] [PubMed] [Google Scholar]
15.Rha B., Dahl R.M., Moyes J., et al. Performance of surveillance case definitions in detecting respiratory syncytial virus infection among young children hospitalized with severe respiratory illness - South Africa, 2009-2014. J Pediatric Infect Dis Soc. 2019;8(4):325–333. doi: 10.1093/jpids/piy055. [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Subissi L., Bossuyt N., Reynders M., et al. Capturing respiratory syncytial virus season in Belgium using the influenza severe acute respiratory infection surveillance network, season 2018/19. Euro Surveill. 2020;25(39) doi: 10.2807/1560-7917.ES.2020.25.39.1900627. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Nyawanda B.O., Mott J.A., Njuguna H.N., et al. Evaluation of case definitions to detect respiratory syncytial virus infection in hospitalized children below 5 years in Rural Western Kenya, 2009-2013. BMC Infect Dis. 2016;16(1) doi: 10.1186/s12879-016-1532-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Davis W., Duque J., Huang Q.S., et al. Sensitivity and specificity of surveillance case definitions in detection of influenza and respiratory syncytial virus among hospitalized patients, New Zealand, 2012–2016. J Infect. 2022;84(2):216–226. doi: 10.1016/j.jinf.2021.12.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Murray E.L., Khagayi S., Ope M., et al. What are the most sensitive and specific sign and symptom combinations for influenza in patients hospitalized with acute respiratory illness? Results from western Kenya, January 2007-July 2010. Epidemiol Infect. 2013;141(1):212–222. doi: 10.1017/S095026881200043X. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.World Health Organization (WHO) 2024. RSV surveillance case definitions.https://www.who.int/teams/global-influenza-programme/global-respiratory-syncytial-virus-surveillance/case-definitions [Google Scholar]
21.Harris P.A., Taylor R., Thielke R., Payne J., Gonzalez N., Conde J.G. Research electronic data capture (REDCap)-A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. doi: 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
22.von Elm E., Altman D.G., Egger M., Pocock S.J., Gøtzsche P.C., Vandenbroucke J.P. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–349. doi: 10.1016/j.jclinepi.2007.11.008. [DOI] [PubMed] [Google Scholar]
23.Gill P.J., Thavam T., Anwar M.R., Zhu J., To T., Mahant S. Pediatric clinical classification system for use in Canadian inpatient settings. PLoS One. 2022;17(8) doi: 10.1371/journal.pone.0273580. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Elnifro E.M., Ashshi A.M., Cooper R.J., Klapper P.E. Multiplex PCR: optimization and application in diagnostic virology. Clin Microbiol Rev. 2000;13(4):559–570. doi: 10.1128/CMR.13.4.559. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Robin X., Turck N., Hainard A., et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinf. 2011;12 doi: 10.1186/1471-2105-12-77. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Woodruff R.C., Campbell A.P., Taylor C.A., et al. Risk factors for severe COVID-19 in children. Pediatrics. 2022;149(1) doi: 10.1542/peds.2021-053418. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Thorburn K. Pre-existing disease is associated with a significantly higher risk of death in severe respiratory syncytial virus infection. Arch Dis Child. 2009;94(2):99–103. doi: 10.1136/adc.2008.139188. [DOI] [PubMed] [Google Scholar]
28.Havers F., Fry A.M., Chen J., et al. Hospitalizations attributable to respiratory infections among children with neurologic disorders. J Pediatr. 2016;170:135–141.e5. doi: 10.1016/j.jpeds.2015.11.030. [DOI] [PubMed] [Google Scholar]
29.Kompaniyets L., Agathis N.T., Nelson J.M., et al. Underlying medical conditions associated with severe COVID-19 Illness among Children. JAMA Netw Open. 2021;4(6) doi: 10.1001/jamanetworkopen.2021.11182. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Debiaggi M., Canducci F., Ceresola E.R., Clementi M. The role of infections and coinfections with newly identified and emerging respiratory viruses in children. Virol J. 2012;9 doi: 10.1186/1743-422X-9-247. [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Hirve S., Crawford N., Palekar R., et al. Clinical characteristics, predictors, and performance of case definition—interim results from the WHO global respiratory syncytial virus surveillance pilot. Influenza Other Respir Viruses. 2020;14(6):647–657. doi: 10.1111/irv.12688. [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Li J., Song C.L., Wang T., et al. Etiological and epidemiological characteristics of severe acute respiratory infection caused by multiple viruses and Mycoplasma pneumoniae in adult patients in Jinshan, Shanghai: a pilot hospital-based surveillance study. PLoS One. 2021;16(3) doi: 10.1371/journal.pone.0248750. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Figs. S1 and S2 and Tables S1–S19

mmc1.docx^{(508.2KB, docx)}

[bib1] 1.Kyu H.H., Vongpradith A., Sirota S.B., et al. Age–sex differences in the global burden of lower respiratory infections and risk factors, 1990–2019: results from the Global Burden of Disease Study 2019. Lancet Infect Dis. 2022;22(11):1626–1647. doi: 10.1016/S1473-3099(22)00510-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib2] 2.Brady M., Duffy R., Domegan L., et al. Establishing severe acute respiratory infection (SARI) surveillance in a sentinel hospital, Ireland, 2021 to 2022. Euro Surveill. 2023;28(23) doi: 10.2807/1560-7917.es.2023.28.23.2200740. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib3] 3.Perin J., Mulick A., Yeung D., et al. Global, regional, and national causes of under-5 mortality in 2000–19: an updated systematic analysis with implications for the sustainable development goals. Lancet Child Adolesc Health. 2022;6(2):10–115. doi: 10.1016/S2352-4642(21)00311-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib4] 4.Williams B.G., Gouws E., Boschi-Pinto C., Bryce J., Dye C. Estimates of world-wide distribution of child deaths from acute respiratory infections. Lancet Infect Dis. 2002;2(1):25–32. doi: 10.1016/S1473-3099(01)00170-0. [DOI] [PubMed] [Google Scholar]

[bib5] 5.Fitzner J., Qasmieh S., Mounts A.W., et al. Revision of clinical case definitions: influenza-like illness and severe acute respiratory infection. Bull World Health Organ. 2018;96(2):122–128. doi: 10.2471/BLT.17.194514. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib6] 6.Rowlinson E., Peters L., Mansour A., et al. Comparison of common acute respiratory infection case definitions for identification of hospitalized influenza cases at a population-based surveillance site in Egypt. PLoS One. 2021;16(3) doi: 10.1371/journal.pone.0248563. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib7] 7.Institute of Medicine (US) Forum on Microbial Threats . National Academies Press; 2007. Global infectious disease surveillance and detection. [DOI] [PubMed] [Google Scholar]

[bib8] 8.Khamphaphongphane B., Chiew M., Mott J.A., et al. Estimating the national burden of hospitalizations for influenza-associated severe acute respiratory infection in the Lao People’s Democratic Republic, 2016. West Pac Surveill Response J. 2021;12(2) doi: 10.5365/wpsar.2020.11.2.001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib9] 9.Makokha C., Mott J., Njuguna H.N., et al. Comparison of severe acute respiratory illness (sari) and clinical pneumonia case definitions for the detection of influenza virus infections among hospitalized patients, western Kenya, 2009-2013. Influenza Other Respir Viruses. 2016;10(4):333–339. doi: 10.1111/irv.12382. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] 10.Galanti M., Birger R., Ud-Dean M., et al. Longitudinal active sampling for respiratory viral infections across age groups. Influenza Other Respir Viruses. 2019;13(3):226–232. doi: 10.1111/irv.12629. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib11] 11.Tsagarakis N.J., Sideri A., Makridis P., Triantafyllou A., Stamoulakatou A., Papadogeorgaki E. Age-related prevalence of common upper respiratory pathogens, based on the application of the Film Array Respiratory panel in a tertiary hospital in Greece. Medicine. 2018;97(22) doi: 10.1097/MD.0000000000010903. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib12] 12.Mandelia Y., Procop G.W., Richter S.S., Worley S., Liu W., Esper F. Dynamics and predisposition of respiratory viral co-infections in children and adults. Clin Microbiol Infection. 2021;27(4):631.e1–631.e6. doi: 10.1016/j.cmi.2020.05.042. [DOI] [PubMed] [Google Scholar]

[bib13] 13.Klink T., Rankin D.A., Piya B., et al. Evaluating the diagnostic accuracy of the WHO severe acute respiratory infection (SARI) criteria in Middle Eastern children under two years over three respiratory seasons. PLoS One. 2020;15(4) doi: 10.1371/journal.pone.0232188. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib14] 14.Amini R., Gilca R., Douville-Fradet M., Boulianne N., De Serres G. Evaluation of the new world health organization case definition of severe acute respiratory infection for influenza surveillance during the peak weeks of two influenza seasons in Quebec, Canada. J Pediatric Infect Dis Soc. 2017;6(3):297–300. doi: 10.1093/jpids/piw044. [DOI] [PubMed] [Google Scholar]

[bib15] 15.Rha B., Dahl R.M., Moyes J., et al. Performance of surveillance case definitions in detecting respiratory syncytial virus infection among young children hospitalized with severe respiratory illness - South Africa, 2009-2014. J Pediatric Infect Dis Soc. 2019;8(4):325–333. doi: 10.1093/jpids/piy055. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib16] 16.Subissi L., Bossuyt N., Reynders M., et al. Capturing respiratory syncytial virus season in Belgium using the influenza severe acute respiratory infection surveillance network, season 2018/19. Euro Surveill. 2020;25(39) doi: 10.2807/1560-7917.ES.2020.25.39.1900627. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] 17.Nyawanda B.O., Mott J.A., Njuguna H.N., et al. Evaluation of case definitions to detect respiratory syncytial virus infection in hospitalized children below 5 years in Rural Western Kenya, 2009-2013. BMC Infect Dis. 2016;16(1) doi: 10.1186/s12879-016-1532-0. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] 18.Davis W., Duque J., Huang Q.S., et al. Sensitivity and specificity of surveillance case definitions in detection of influenza and respiratory syncytial virus among hospitalized patients, New Zealand, 2012–2016. J Infect. 2022;84(2):216–226. doi: 10.1016/j.jinf.2021.12.012. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib19] 19.Murray E.L., Khagayi S., Ope M., et al. What are the most sensitive and specific sign and symptom combinations for influenza in patients hospitalized with acute respiratory illness? Results from western Kenya, January 2007-July 2010. Epidemiol Infect. 2013;141(1):212–222. doi: 10.1017/S095026881200043X. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib20] 20.World Health Organization (WHO) 2024. RSV surveillance case definitions.https://www.who.int/teams/global-influenza-programme/global-respiratory-syncytial-virus-surveillance/case-definitions [Google Scholar]

[bib21] 21.Harris P.A., Taylor R., Thielke R., Payne J., Gonzalez N., Conde J.G. Research electronic data capture (REDCap)-A metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform. 2009;42(2):377–381. doi: 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib22] 22.von Elm E., Altman D.G., Egger M., Pocock S.J., Gøtzsche P.C., Vandenbroucke J.P. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–349. doi: 10.1016/j.jclinepi.2007.11.008. [DOI] [PubMed] [Google Scholar]

[bib23] 23.Gill P.J., Thavam T., Anwar M.R., Zhu J., To T., Mahant S. Pediatric clinical classification system for use in Canadian inpatient settings. PLoS One. 2022;17(8) doi: 10.1371/journal.pone.0273580. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] 24.Elnifro E.M., Ashshi A.M., Cooper R.J., Klapper P.E. Multiplex PCR: optimization and application in diagnostic virology. Clin Microbiol Rev. 2000;13(4):559–570. doi: 10.1128/CMR.13.4.559. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib25] 25.Robin X., Turck N., Hainard A., et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinf. 2011;12 doi: 10.1186/1471-2105-12-77. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] 26.Woodruff R.C., Campbell A.P., Taylor C.A., et al. Risk factors for severe COVID-19 in children. Pediatrics. 2022;149(1) doi: 10.1542/peds.2021-053418. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib27] 27.Thorburn K. Pre-existing disease is associated with a significantly higher risk of death in severe respiratory syncytial virus infection. Arch Dis Child. 2009;94(2):99–103. doi: 10.1136/adc.2008.139188. [DOI] [PubMed] [Google Scholar]

[bib28] 28.Havers F., Fry A.M., Chen J., et al. Hospitalizations attributable to respiratory infections among children with neurologic disorders. J Pediatr. 2016;170:135–141.e5. doi: 10.1016/j.jpeds.2015.11.030. [DOI] [PubMed] [Google Scholar]

[bib29] 29.Kompaniyets L., Agathis N.T., Nelson J.M., et al. Underlying medical conditions associated with severe COVID-19 Illness among Children. JAMA Netw Open. 2021;4(6) doi: 10.1001/jamanetworkopen.2021.11182. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib30] 30.Debiaggi M., Canducci F., Ceresola E.R., Clementi M. The role of infections and coinfections with newly identified and emerging respiratory viruses in children. Virol J. 2012;9 doi: 10.1186/1743-422X-9-247. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib31] 31.Hirve S., Crawford N., Palekar R., et al. Clinical characteristics, predictors, and performance of case definition—interim results from the WHO global respiratory syncytial virus surveillance pilot. Influenza Other Respir Viruses. 2020;14(6):647–657. doi: 10.1111/irv.12688. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib32] 32.Li J., Song C.L., Wang T., et al. Etiological and epidemiological characteristics of severe acute respiratory infection caused by multiple viruses and Mycoplasma pneumoniae in adult patients in Jinshan, Shanghai: a pilot hospital-based surveillance study. PLoS One. 2021;16(3) doi: 10.1371/journal.pone.0248750. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Performance of the World Health Organization (WHO) severe acute respiratory infection (SARI) case definitions in hospitalized children and youth: cross-sectional study

Peter J Gill

Caitlyn L Kaziev

Haifa Mtaweh

Tuana Kant

Claire Seaton

Daniel S Farrar

Hayley Wagman

Mei Han

Rohini R Datta

Sanjay Mahant

Gabrielle Freire

Aaron Campigotto

Jeffrey N Bone

Manish Sadarangani

Francine Buchanan

Shaun K Morris

Summary

Background

Methods

Findings

Interpretation

Funding

Research in context.

Evidence before this study

Added value of this study

Implications of all the available evidence

Introduction

Methods

Study design

Study population

Data collection

Severe acute respiratory infection (SARI) case definitions

Data analysis

Role of the funding source

Results

Description of study population

Table 1.

Clinical features

Table 2.

WHO SARI case definition

Table 3.

Table 4.

WHO SARI case definition by age and virus

Modified SARI case definitions

Fig. 1.

Discussion

Conclusion

Contributors

Data sharing statement

Declaration of interests

Acknowledgements

Footnotes

Contributor Information

Appendix ASupplementary data

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases