RESEARCH ARTICLE
Global Role and Burden of Influenza in
Pediatric Respiratory Hospitalizations, 1982–
2012: A Systematic Analysis
Kathryn E. Lafond
1,2
*, Harish Nair
3,4
, Mohammad Hafiz Rasooly
5
, Fátima Valente
6
,
Robert Booy
7
, Mahmudur Rahman
8
, Paul Kitsutani
1
, Hongjie Yu
9
, Guiselle Guzman
10
,
Daouda Coulibaly
11
, Julio Armero
12
, Daddi Jima
13
, Stephen R. C. Howie
14,15,16
,
William Ampofo
17
, Ricardo Mena
18
, Mandeep Chadha
19
, Ondri Dwi Sampurno
20
, Gideon
O. Emukule
21
, Zuridin Nurmatov
22
, Andrew Corwin
1
, Jean Michel Heraud
23
, Daniel
E. Noyola
24
, Radu Cojocaru
25
, Pagbajabyn Nymadawa
26
, Amal Barakat
27
,
Adebayo Adedeji
28
, Marta von Horoch
29
, Remigio Olveda
30
, Thierry Nyatanyi
31
,
Marietjie Venter
32,33,34
, Vida Mmbaga
35
, Malinee Chittaganpitch
36
, Tran Hien Nguyen
37
,
Andros Theo
38
, Melissa Whaley
1
, Eduardo Azziz-Baumgartner
1
, Joseph Bresee
1
,
Harry Campbell
3
, Marc-Alain Widdowson
1
*, Global Respiratory Hospitalizations—
Influenza Proportion Positive (GRIPP) Working Group
¶
1 Influenza Division, Centers for Disease Control and Prevention, Atlanta, Georgia, United States of
America, 2 School of Health Sciences, University of Tampere, Tampere, Finland, 3 Centre for Global Health
Research, University of Edinburgh, Edinburgh, United Kingdom, 4 Public Health Foundation of India, New
Delhi, India, 5 Afghanistan National Public Health Institute, Ministry of Public Health, Kabul, Afghanistan,
6 National Directorate of Public Health, Ministry of Health, Luanda, Angola, 7 National Centre for
Immunisation Research and Surveillance, The Children’s Hospital at Westmead, Westmead, New South
Wales, Australia, 8 Institute of Epidemiology, Disease Control and Research, Dhaka, Bangladesh,
9 Division of Infectious Disease, Key Laboratory of Surveillance and Early-warning on Infectious Disease,
Chinese Centre for Disease Control and Prevention, Beijing, China, 10 Caja Costarricense de Seguro
Social, San José, Costa Rica, 11 Pasteur Institut of Côte d’Ivoire, Abidjan, Côte d’ Ivoire, 12 Ministerio de
Salud de El Salvador, San Salvador, El Salvador, 13 Ethiopian Public Health Institute, Addis Ababa,
Ethiopia, 14 Medical Research Council Unit, Fajara, The Gambia, 15 Department of Paediatrics, University
of Auckland, Auckland, New Zealand, 16 Centre for International Health, University of Otago, Dunedin, New
Zealand, 17 Noguchi Memorial Institute for Medical Research, University of Ghana, Accra, Ghana,
18 Ministerio de Salud Publica y Asistencia Social, Guatemala City, Guatemala, 19 National Institute of
Virology, Pune, India, 20 National Institute of Health Research and Development, Jakarta, Indonesia,
21 Centers for Disease Control and Prevention, Nairobi, Kenya, 22 Ministry of Health, Bishkek, Kyrgyzstan,
23 National Influenza Centre, Virology Unit, Institut Pasteur of Madagascar, Antananarivo, Madagascar,
24 Universidad Autónoma de San Luis Potosí, San Luis Potosí, Mexico, 25 National Centre for Public
Health, Chisinau, Republic of Moldova, 26 National Influenza Center, Ulaanbaatar, Mongolia, 27 Institut
National d’Hygiène, Ministère de la Santé, Rabat, Morocco, 28 Federal Ministry of Health, Abuja, Nigeria,
29 Ministerio de Salud Publica y Bienestar Social, Asunción, Paraguay, 30 Research Institute for Tropical
Medicine, Manila, Philippines, 31 Ministry of Health, Kigali, Rwanda, 32 National Institute for Communicable
Diseases, Johannesburg, South Africa, 33 Zoonoses Research Unit, Department Medical Virology,
University of Pretoria, Pretoria, South Africa, 34 Division of Global Health Protection, Centers for Disease
Control and Prevention, Atlanta, Georgia, United States of America, 35 Ministry of Health, Dar es Salaam,
Tanzania,
36 National Institute of Health, Ministry of Public Health, Nonthaburi, Thailand, 37 National
Institute of Hygiene and Epidemiology, Hanoi, Viet Nam, 38 Virology Laboratory, University Teaching
Hospital, Lusaka, Zambia
¶ Membership of the GRIPP Working Group is provided in the Acknowledgments.
*
gmj3@cdc.gov (KEL); zux5@cdc.gov (MAW)
PLOS Medicine | DOI:10.1371/journal.pmed.1001977 March 24, 2016 1/19
OPEN ACCESS
Citation: Lafond KE, Nair H, Rasooly MH, Valente F,
Booy R, Rahman M, et al. (2016) Global Role and
Burden of Influenza in Pediatric Respiratory
Hospitalizations, 1982–2012: A Systematic Analysis.
PLoS Med 13(3): e1001977. doi:10.1371/journal.
pmed.1001977
Academic Editor: James K. Tumwine, Makerere
University Medical School, UGANDA
Received: March 3, 2015
Accepted: February 5, 2016
Published: March 24, 2016
Copyright: This is an open access article, free of all
copyright, and may be freely reproduced, distributed,
transmitted, modified, built upon, or otherwise used
by anyone for any lawful purpose. The work is made
available under the
Creative Commons CC0 public
domain dedication.
Data Availability Statement: All relevant data are
provided within the paper and its Supporting
Information files.
Funding: Funding for this study was provided
entirely by the U.S. Centers for Disease Control and
Prevention (CDC). The study was designed by the
authors, and the results and conclusions do not
necessarily reflect the official position of the CDC.
Competing Interests: We have read the journal's
policy and have the following competing interests:
DEN has participated on an influenza advisory board
Abstract
Background
The global burden of pediatric severe respiratory illness is substantial, and influenza viruses
contribute to this burden. Systematic surveillance and testing for influenza among hospital-
ized children has expanded globally over the past decade. However, only a fraction of the
data has been used to estimate influenza burden. In this analysis, we use surveillance data
to provide an estimate of influenza-associated hospi talizations among children worldwide.
Methods and Findings
We aggregated data from a systematic review (n=108) and surveillance platforms (n=37)
to calculate a pooled estimate of the proportion of samples collected from children hospital-
ized with respiratory illnesses and positive for influenza by age group (<6 mo, <1y,<2y,<5
y, 5–17 y, and <18 y). We applied this proportion to global estimates of acute lower respira-
tory infection hospitalizations among children aged <1 y and <5 y, to obtain the number and
per capita rate of influenza-associated hospitalizations by geographic region and socio-eco-
nomic status.
Influenza was associated with 10% (95% CI 8%–11%) of respiratory hospitalizations in
children <18 y worldwide, ranging from 5% (95% CI 3%–7%) among children <6 mo to 16%
(95% CI 14%–20%) among children 5–17 y. On average, we estimated that influenza
results in approximately 374,000 (95% CI 264,000 to 539,000) hospitalizations in children
<1y— of which 228,000 (95% CI 150,000 to 344,000) occur in children <6mo—and
870,000 (95% CI 610,000 to 1,23 7,000) hospitalizations in children <5 y annually. Influ-
enza-associated hospitalization rates were more than three times higher in developing
countries than in industrialized countries (150/100,000 children/year versus 48/100,000).
However, differences in hospitalization practices between settings are an important limita-
tion in interpreting these findings.
Conclusions
Influenza is an important contributor to respiratory hospitalizations among young children
worldwide. Increasing influenza vaccination coverage among young children and pregnant
women could reduce this burden and protect infants <6 mo.
Introduction
Influenza virus infections are a substantial contributor to respiratory morbidity and mortality,
with the highest burden of severe disease experienced by those aged <5 y and 65 y [
1–3].
Until recently, however, estimates of influenza burden have been largely obtained from
resource-rich settings with temperate climates with clearly defined influenza sea sonality, with
few estimates available from lower-income countries [
4–6]. This lack of information on influ-
enza burden in resource-limited settings has hampered informed consideration of implementa-
tion of preventive measures such as vaccination.
In the last 10 y, however, the global expansion of influenza surveillance and laboratory
capacity for influenza testing by reverse transcription PCR has led to dramatic increases in
Global Pediatric Hospitalizations for Influenza
PLOS Medicine | DOI:10.1371/journal.pmed.1001977 March 24, 2016 2/19
for Novartis. RB works with all major manufacturers
of influenza vaccines in an advisory capacity, as a
researcher on vaccines and as presenter of
academic info at conferences, receiving support to
travel and attend such conferences. The authors
have declared that no other competing interests exist.
Abbreviations: ALRI, acute lower respiratory
infection; GRIPP, Global Respiratory
Hospitalizations–Influenza Proportion Positive; IQR,
interquartile range; RSV, respiratory syncytial virus;
SARI, severe acute respiratory illness.
testing in settings with previously sparse data. Many countries worldwide now perform hospi-
tal-based influenza surveillance among patients hospitalized with severe acute respiratory ill-
ness (SARI) [
7–9]. These systems have prove n useful in tracking influenza activity among
hospitalized patients, but only a fraction of the collected data has been used to estimate the bur-
den of influenza-associated hospitalizations. A 2011 meta-analysis using data from 16 popula-
tion-based research sites and published literature estimated that there were 1 million cases of
influenza-associated SARI episodes globally in children aged <5 y old in 2008. In this study,
the per capita rate of severe influenza illness in developing countries was at least double that of
industrialized countries [6]. This rate-based approach was mainly derived from sites with a
limited population at risk under surveillance, and therefore represented only a small fraction of
hospitals conducting surveillance for severe influenza disease globally.
We aimed to update the previous influenza burden estimat es in young children, making full
use of updated and expan ded surveillance data from the past decade (2003–2012), both before
and after the 2009 pandemic influenza emergence. We further aimed to extend previous studies
by providing global estimates of the prevalence of influenza among acute lower respiratory
infection (ALRI) hospitalizations among both younger (<5 y) and older (5–17 y) children.
Methods
We aggregated data from all eligible published etiologic studies of influenza-associated respira-
tory illness among hospitalized children, which we supplemented with data from a working
group of inpatient surveillance systems worldwide. We then calculated a final pooled estimate
of the proportion of tested samples that were positive for influenza by reverse transcription
PCR among children aged <18 y (referred to as proportion or percent positive), using age-
group-specific random-effects log-binomial regression models. Finally, we applied the aggre-
gate pooled proportion positive among children <1 and <5 y to age-specific denominators of
global hospitalizations for ALRI among these two age groups [10] to obtain the number and
rate (per 100,000 children per year) of pediatric influenza-associated hospitalizations, by
World Health Organization (WHO) region and Unite d Nations (UN) country development
status.
Systematic Review of the Literature
We searched nine online databases (PubMed, Embase, Web of Science, CINAHL [Cumulative
Index to Nursing and Allied Health Literature], IndMed, LILACS [Literatura Latino -Ameri-
cana e do Caribe em Ciências da Saúde], WHOLIS [WHO Library Database], CNKI [China
National Knowledge Infrastructure], and the Global Health Database) to identify article s pub-
lished from 1 January 1996 to 1 June 2012. The search was conducted with no language restric-
tions, and full search terms are provided in
S1 Table. Briefly, keywords included “influenza” or
“viral etiology” and other designators of respiratory illness such as “acute respiratory infection”
and “influenza-like illness”. Searches via the CNKI Chinese-language database were conducted
by native Mandarin speakers.
Identified articles were screened by two independent reviewers (two from K. E. L., M. W., E.
A.-B., M.-A. W., P. Glew, S. Mei, Z. Suizan) for inclusion in the analysis, and duplicates were
removed. The inclusion criteria were as follows: (1) original study with human participants, (2)
laboratory testing for influenza, with description of the type of diagnostic method used, (3)
minimum of 12 mo of continuous surveillance, (4) specified case definition (such as ALRI,
SARI, or acute respiratory illness) or other clear criteria for specimen collection and testing, (5)
hospitalized patients (excluding nosocomial infections), (6) number of enrolled cases from
whom clinical specimens were collected and found positive was provided, and (7) minimum of
Global Pediatric Hospitalizations for Influenza
PLOS Medicine | DOI:10.1371/journal.pmed.1001977 March 24, 2016 3/19
50 children (<18 y or “pediatric” as defined by authors) tested for influenza, in order to screen
out small, potentially unreliable studies from the study dataset. For title and abstracts that met
these criteria, full-text articles were obtained and re-screened. Full-text articles written in lan-
guages other than English were screened twice by co-investigators who could read the relevant
language. Any discrepancies were discussed and resolved by reviewers. Independent screening
was concordant for 92% of full-text articles, with 100% concordance after joint discussion of
discrepancies.
Key data from each eligible article were abstracted by two independent reviewers. Data
abstracted included the total number of inpatients tested and total positive for influenza by age
group and year, case definition and diagnostic test, WHO region, World Bank income level
(low, lower-middle, upper-middle, or high income) [
11], and UN country development status
(industrialized or developing) [
12].
Quality Assessment
Data quality for each eligible article was scored using a modified Newcastle–Ottawa checklist
for bias assessment [
13], with a score of zero or one for each of the following sources of bias:
sampling process (explicit description of the sampling process for enrollment), case definition
(specificity of enrollment criteria), and outcome (clarity of reported results). We explored the
association between quality score and the percent positive using rank-sum non-parametric
tests among all eligible articles included in the pooled analysis.
Surveillance Data
To supplement data from publish ed studies, we compiled data from surveillance platforms that
conducted hospital-based influenza surveillance. We established a working group, the Global
Respiratory Hospitalizations–Influenza Proportion Positive (GRIPP) Working Group. To be
eligible, surveillance platforms needed to conduct systematic year-round inpatient enrollment,
with testing for at least 12 mo and >50 pediatric patients. Forty-eight partners were contacted,
of which 37 had eligible data and agreed to participate. Data were collected using a standard
format. Variables included the number of persons tested and positive for influenza by calendar
year and age group, as well as surveillance system information such as the total number of
inpatient sites and case definition used. If surveillance data were also represented in a report
identified through the systematic review, the more detailed working group dataset was used,
and the published article was excluded as a duplicate source.
Statistical Analysis
We first described the median proportion positive by age group, study duration, calendar year,
number tested, diagnostic method, case definition, study population, WHO region, and coun-
try income level among all eligible datasets. We found substantial variation by age, diagnostic
test, and calendar year. To reduce the influence of data from less reliable diagnostic methods
(such as immunofluorescence [low sensitivity] and single serological samples [higher likeli-
hood of false-positive findings]), we restricted the data for the meta-analysis to sites that uti-
lized PCR for diagnosis, which is the accepted diagnostic gold standard. Further, to ensure that
pooled estimates reflected seasonal rather than pandemic years, we excluded data from 2009.
Pooled estimates of the proportion of respiratory hospitalizations due to influenza were cal-
culated by mixed regression model for each of the following age groups: <6 mo, <1y,<2y,
<5y,5–17 y, and <18 y. If a dataset used age ranges that did not line up with our definitions, it
was included in the smallest range that contained both bounds (e.g., a dataset from children
0–36 mo was included in the <5 y analysis). The same datasets could provide estimates for
Global Pediatric Hospitalizations for Influenza
PLOS Medicine | DOI:10.1371/journal.pmed.1001977 March 24, 2016 4/19
multiple age groups if they provided number tested and positive for influenza for each age
group.
The mixed regression model in SAS version 9.3 (SAS Institute) used a log-linked binomial
distribution of input values, and included the number tested and positive for influenza by data-
set and calendar year for each age group. If a single dataset provided data by year, then each
year was treated as a single observation in the model, and the dataset was defined as a cluster.
Random effects were accounted for at the dataset level, irrespective of the number of observa-
tions within the dataset.
We then applied our pooled proportion positive for influenza viruses to global estimates of
the total annual number of hospitalizations for ALRI among children <5 y and <1 y from Nair
et al. [
10] and adjusted for 2012 population to calculate the total number of influenza-associ-
ated hospitalizations in these two age groups (for which denominators were available). For
children <5 y, we also applied stratified pooled estimates (by WHO region and UN country
development status) to the appropriate ALRI hospitalization denominators. We then divided
these total numbers of influenza-associated hospitalizations by total age-specific population (at
global and regional levels, and separately for developing and industrialized countries) to esti-
mate annual per capita influenza-associated hospitalization episodes [
14].
No estimates of the frequency of ALRI among children under 6 mo existed to which we
could apply the proporti on positive for influenza viruses. Therefore, we calculated an incidence
rate ratio of influenza-associated ALRI bet ween children aged 0–5 mo and 6–11 mo, collected
as part of a separate study [
6](S2 Appendix). We applied this ratio to our estimate of the num-
ber of influ enza-associated ALRI hospitalizations in children under 1 y, assuming uniform dis-
tribution of the population under 1 y, to estimate the number of influenza-associated
hospitalizations among children aged 0–5 mo.
Results
The systematic literature search identified 38,006 unique records from the nine scientific litera-
ture databases, of which 957 full-text articles were reviewed, and 108 included in the descriptive
analysis (
Fig 1) (full list of included articles provided in S1 Appendix). In addition, 37 surveil-
lance datasets for periods ranging from 1 to 7 y, each with data from 1–49 inpatient facilities,
were provided by the GRIPP Working Group. Combined, the literature search and working
group resulted in a total of 145 unique data sources from 350 sites in 60 countries across all
WHO regions, including southern and northern hemisphere temperate climates as well as
tropical and arid regions of Asia and Africa. The two data sources combined covered a 31-y
period from 1982 to 2012. More than half (55/108, 51%) of the published studies had a 3/3
quality score. Compared to published articles, surveillance datasets were more recently col-
lected (median start year 2009 versus 2002), of longer duration (median duration 3 y versus 2
y), more likely to be PCR-based (41% versus 28%), and more likely to use the SARI case defini-
tion (84% versus 0%). Additionally, surveillance datasets were also more commonly from low-
income countries compared to published articles (41% versus 6%, respectively;
Table 1).
The crude median percent of respiratory samples that were influenza positive among
patients aged 5 – 17 y was more than double that among those <5 y (15%, interquartile range
[IQR] 10%–22%, versus 6%, IQR 3%–9%, p < 0.001) and was also significantly higher in sur-
veillance data than in publish ed articles (9%, IQR 6%–12%, versus 5%, IQR 3%–9%, p < 0.01)
(
Table 2). The median percent positive was also significantly higher in 2009, when pandemic
influenza A(H1N1)pdm09 virus emerged, and in following years, compared to before 2009
(13% in 2009, IQR 6%–20%, versus 9% after 2009, IQR 5%–11%, and 5% before 2009, IQR
3%–9%, p < 0.001). The majority of the datasets (57%) comprised results from PCR diagnosis.
Global Pediatric Hospitalizations for Influenza
PLOS Medicine | DOI:10.1371/journal.pmed.1001977 March 24, 2016 5/19