Influenza in Children
• JID 2004:190 (15 October) • 1369
M A J O R A R T I C L E
Burden of Influenza in Children in the Community
Terho Heikkinen,1Heli Silvennoinen,1Ville Peltola,1Thedi Ziegler,3,aRaija Vainionpa ¨a ¨,3Tytti Vuorinen,3
Leena Kainulainen,1Tuomo Puhakka,2Tuomas Jartti,1Pia Toikka,1Pasi Lehtinen,1Taina Routi,1and Taina Juve ´n1
2Otorhinolaryngology, Turku University Hospital, and
3Department of Virology, University of Turku, Turku, Finland
pitalized for influenza-attributable illnesses. However, most children with influenza are treated as outpatients, and
scarce data are available on the burden of influenza in these children.
We performed a prospective study of respiratory infections in preenrolled cohorts of children ?13
years old during 2 consecutive respiratory seasons (2231 child-seasons of follow-up). At any sign of respiratory
infection, we examined the children and obtained a nasal swab for the detection of influenza. The parents filled
out daily symptom diaries. Of all the enrollees, 94% remained active participants in the study.
The average annual rate of influenza was highest (179 cases/1000 children) among children !3 years
old. Acute otitis media developed as a complication of influenza in 39.7% of children !3 years old. For every 100
influenza-infected children !3 years old, there were 195 days of parental work loss (mean duration, 3.2 days).
Influenza causes a substantial burden of illness on outpatient children and their families. Vac-
cination of children !3 years old might be beneficial for reducing the direct and indirect costs of influenza in
Influenza vaccination of healthy children is encouraged because children are frequently hos-
Annual morbidity due to influenza is highest in chil-
dren, in whom the attack rates are often 130% [1–4].
Children are also considered to be the main dissemi-
nators of influenza in the community [4–9]. Recent
studies have documented that infants and young chil-
dren without underlying medical conditions are hos-
pitalized for influenza-attributable illnesses at ratesthat
are comparable to those of adults with high-risk con-
ditions [10–13]. The high risk of hospitalizationofchil-
dren has initiated an intensive discussion about the
Received 9 March 2004; accepted 30 April 2004; electronically published 15
Presented in part: Options for the Control of Influenza V, Okinawa, Japan, 7–11
October 2003 (abstract W03-5).
Financial support: Wyeth; GlaxoSmithKline; European Scientific Working Group on
Influenza; Academy of Finland; Foundation for Pediatric Research in Finland; Jenny
and Antti Wihuri Foundation, Finland.
None of the sponsors played any role in the design or conduct of the study; in
the collection, management, analysis, or interpretation of the data; or in the
preparation or review of the manuscript. The decision to publish the data was made
solely by the authors, who did not seek or obtain any sort of authorization for
publication from the sponsors.
Potential conflict of interest: T.H. has served as a consultant to Wyeth and has
received honoraria for lectures at academic meetings organized by Wyeth and
aCurrent affiliation: National Public Health Institute, Helsinki, Finland.
Hospital, FIN-20520 Turku, Finland (firstname.lastname@example.org).
The Journal of Infectious Diseases
? 2004 by the Infectious Diseases Society of America. All rights reserved.
universal vaccination of children against influenza[10–
14]. In 2002, the US Advisory Committee on Immu-
nization Practices decided to encourage the vaccination
of all children 6–23 months old, when feasible ,
and, in October 2003, the Committee voted to rec-
ommend that all children in this age group be vacci-
nated annually against influenza, unless contraindica-
tions are present .
The cocirculation of other viruses, particularly res-
piratory syncytial virus, during influenza epidemics is
a serious obstacle to reliable estimations of the burden
of influenza in children. Accordingly, there are doubts
about whether influenza is responsible for all of the
excess morbidity that is attributed to it . Moreover,
hospitalized children represent only a small proportion
of the overall incidence of pediatric influenza. Most
children with influenza are treated as outpatients, and
it is probable that these cases account for the greater
part of the total disease burden in children. The wide-
spread vaccination of healthy children againstinfluenza
will likely requireanassessmentofthecost-effectiveness
of the intervention, and an estimation of the direct and
indirect costs of influenza and its complications in chil-
dren treated as outpatients is of crucial importance in
The present prospective study was specifically de-
signed to determine the total burden of influenza in
by guest on November 3, 2015
1370 • JID 2004:190 (15 October) • Heikkinen et al.
season of follow-up.
Characteristics of the children at the beginning of each
(n p 1338)
(n p 893)
6 months to !3 years
3 to !7 years
Method of child care
Family day care
Day care center
History of wheezing
Diagnosis of asthma
Influenza vaccination for the season
Data are no. (%) of children.
children in the community. To enable a reliable estimation of
the impact of influenza, we obtained samples for viraldetection
during every episode of respiratory illness in the children.
SUBJECTS AND METHODS
centers, family day care, and schools in our area. All children
!13 years old were eligible for participation. Of 1458 children
initially recruited, 1338 were monitored during the respiratory
season of 2000–2001. For the 2001–2002 season, the study
group consisted of 907 children, of whom 893 were eventually
monitored. Overall, the study comprised 2231 child-seasons of
follow-up. The baseline demographics of the children are pre-
sented in table 1.
The study protocol was approved by the Ethics Committee
of Turku University Hospital. Written, informed consent was
obtained from the parents of all participating children.
The study was performed from 9 October
2000 through 20 May 2001 and from 1 October 2001 through
19 May 2002. The parents were asked to bring their children to
the study clinic whenever fever or signs of respiratory infection
appeared. The study clinic was open every day. At each visit, the
children were examined by a study physician. Chest or sinus
radiographs were routinely obtained for all children who were
clinically suspected of having pneumonia or sinusitis. In addi-
tion to pneumatic otoscopy, tympanometry and spectral-gra-
dient acoustic reflectometry were used to aid in diagnosing
acute otitis media (AOM). Children without any complica-
tions were routinely reexamined after 5–7 days and whenever
the parents deemed it necessary. All visits were free of charge.
Influenza antiviral drugs and rapid diagnostic tests were not
used for any children.
During each episode of respiratory infec-
tion, a nasal swab was obtained from a depth of 2–3 cm in the
nostril by use of a sterile cotton swab that was then inserted
into a vial that contained viral transport medium . Detec-
tion of influenza viruses in the specimens was based on viral
culture in Madin-Darbey canine kidney cells and subsequent
immunoperoxidase staining with monoclonalantibodies,asde-
scribed elsewhere .
During each season, the parents were
provided with 2 daily symptom diaries that consisted of charts
inquiring about the symptoms and absences from day care or
school of the child and from work, school, or day care of other
family members. The days of absenteeism included only actual
days lost; days of illness occurring during free weekends or
other days off were not recorded as causing absenteeism. Pa-
rental compliance with the diaries was good—at least 1 diary
was returned for 97% of theactiveparticipants,andbothdiaries
were returned for 85% of the active participants.
AOM was diagnosed by signs of inflammation
We recruited the participants through day care
of the tympanic membrane, the presence of middle-eareffusion
as detected by pneumatic otoscopy, and ?1 sign of acute in-
fection. The diagnoses of pneumonia and sinusitis were based
on radiological confirmation of the condition. Any complica-
tions were considered to be associated with influenza if they
were diagnosed within 14 days after the clinical visit at which
the influenza-positive specimen was obtained.
groups was done by the x2test (SigmaStat version 2.0; SPSS).
Rates of influenza illnesses.
enza in the children were documented (table 2). For both sea-
sons combined, the annual rates of influenza illness were 179
cases/1000 children !3 years old, 175 cases/1000 children 3–6
years old, and 142 cases/1000 children ?7 years old. Of the
372 influenza illnesses, 301 (81%) were caused by influenza A
viruses and 59 (16%) by influenza B viruses; in 12 cases (3%),
the viruses remained untyped.
Complications, antibiotic treatments, and hospitalizations.
Two of 372 influenza-positive children were excluded from
further analyses because of double viral infection. AOM was
the most frequently diagnosed complication, and it occurred
in 39.7% of children !3 years old (table 3). Pneumonia was
diagnosed in 2.4% and sinusitis was diagnosed in 3.5% of all
A total of 372 episodes of influ-
by guest on November 3, 2015
Influenza in Children • JID 2004:190 (15 October) • 1371
Rates of culture-confirmed influenza in children during 2 consecutive seasons
Season, age group
Rate of influenza cases/
1000 children (95% CI)
Both seasons combined
76 160 (127–193)
CI, confidence interval.
ture-confirmed influenza illness.
Complications, antibiotic treatments, and hospitalizations in 370 children with cul-
(n p 370)
(n p 131)
(n p 148)
(n p 91)
Acute otitis media
Referral to emergency department
cThe indications for antibiotic treatment were acute otitis media (81 cases), pneumonia (9), sinusitis (11),
tonsillitis (1), bronchitis (1), and fever of unknown origin (1).
Data are no. (%) of cases. Data for 2 children with double viral infection were excluded.
, for all comparisons between the age groups (x2test).
, for all comparisons between the age groups (x2test).
children with influenza. Antibiotics were prescribed most often
(42.0%) to children !3 years old.
Three (2.3%) of 131 children !3 years old were referred to
the emergency department because of high fever, inspiratory
stridor, and/or impaired general condition; 1 of these children
was hospitalized with pneumonia. None of the older children
were referred or admitted to hospital because of influenza.
Children’s absences from day care or school.
75.2% of children with influenza were absent from day care or
school for ?1 day (table 4). For every 100 children with in-
fluenza, there were 274 days of absence among children !3
years old, 262 days among children 3–6 years old, and 216 days
among children ?7 years old.
Parental work loss.
A parent missed ?1 day of work be-
cause of the child’s influenza in 61.2% of cases in children !3
years old, 53.9% of cases in children 3–6 years old, and 26.3%
of cases in children ?7 years old (table 4). When calculated
for only those cases in which the child was absent for at least
1 day, the corresponding rates of parental absence from work
in the different age groups were 80.8%, 73.4%, and 33.9%,
respectively. For every 100 children with influenza, there were
195 days of parental work loss for children !3 years old, 146
days for children 3–6 years old, and 54 days for children ?7
Our prospective cohort study has provided new and detailed
evidence for the substantial burden of influenza in children. A
unique feature of this study is that the estimation of the impact
of influenza was not based on epidemiological surveillancedata
but on direct documentation of influenza virus infection on
an individual level. Close clinical follow-up of large preenrolled
cohorts of healthy children throughout 2 consecutive winter
by guest on November 3, 2015
1372 • JID 2004:190 (15 October) • Heikkinen et al.
absence from work because of child’s influenza.
Children’s absenteeism from day care (0–6 years old) or school (7–13 years old) and parental
Category, age group
for ?1 day,
mean (SE), days
Total days of
with influenza (95% CI)
excluded. CI, confidence interval.
aCalculated for children or parents who were absent for at least 1 day.
Data on absenteeism were available for 328 children; 21 children who were cared for at home by a parent were
seasons allowed us to evaluate and compare the impact of
influenza between different age groups of children.
The severity of influenza epidemics varies between different
seasons and viral strains [19, 20]; therefore, results obtained
during 1 epidemic may not be directly generalizable to other
outbreaks. The value of our 2-year follow-up is demonstrated
by the noticeable differences in influenza attack rates between
the different age groups of children during the 2 seasons. In
2000–2001, the outbreak was caused primarily by type A/H1N1
influenza viruses that had not been circulating at epidemic
levels in Finland since 1996 , and the incidence of influenza
was fairly similar in all age groups, although the highest attack
rates were seen in children 3–6 years old. In contrast, during
the 2001–2002 season, when the circulating viruses weremainly
of type A/H3N2, which had been prevalent for several years
before 2000 , the incidence of influenza was 7 times higher
in children !3 years old than in those ?7 years old. It should
be emphasized, however, that both of these influenzaepidemics
in Finland were considered to be mild [21, 22]; therefore, it is
unlikely that our study would have overestimated the impact
of influenza during an average outbreak.
To increase children’s compliance with repeated sampling,
we chose to obtain nasal swabs instead of nasopharyngeal as-
pirates. The sensitivity of nasal swabs for the detection of influ-
enza in children is ∼90%, compared with nasopharyngeal as-
pirates , and it is thus probable that the true incidence of
influenza in our cohort may have been at least 10% higher than
was actually diagnosed. Moreover, although our virus culture
method, which used immunoperoxidase staining with mono-
clonal antibodies, is more sensitive than conventional virus
culture , the use of polymerase chain reaction (PCR) for
the detection of influenza might have furtherincreasedtherates
of influenza in our cohorts . Therefore, our results on the
incidence of influenza should be regarded as conservative es-
timates, and the total burden of influenza in children may be
even substantially higher than we observed. It is most unlikely,
however, that the use of PCR would have affected our findings
regarding the relative rates of complications or average dura-
tions of absenteeism for influenza-infected children.
AOM was the most frequent complication of influenza and
accounted for ∼80% of antibiotic prescriptions in influenza-
infected children. The high incidence of AOM in our cohort
corroborates earlier findings in other settings [12, 24–27].Con-
sidering that the total cost of a single episode of AOM has been
estimated to be US $228–$262 [28–30], it is obvious that the
frequency of this complication makes it one of the key factors
in calculations of the cost-effectiveness of influenzavaccination
Only 1 child in our cohort was hospitalized for an influenza-
associated condition. This finding clearly supports the concept
that the greatest part of the total disease burden of influenza,
including its complications, occurs in the outpatient setting.
However, this finding does not challenge the results of previous
studies that have demonstrated the increased risk of hospital-
ization of children for influenza-attributable illnesses [10–13].
Extrapolation of our data suggests that the annual rate of influ-
enza-associated hospitalization among children !3 years old in
our study would be 135 cases/100,000 children, which is com-
parable to previous estimates from the United States [10–12].
Larger sample sizes and longer follow-up than in the present
study would be needed to determine the rates of rarely occur-
ring severe manifestations of influenza.
Parental work loss for the care of sick children forms a sub-
stantial proportion of the total cost of influenza in children
[31–33]. In our study, both the highest rate and the longest
duration of parental absence from work were observed for chil-
dren !3 years old. By use of a conservative assumption of US
$93 for a day’s salary for a parent in the United States ,
the mean duration of work loss (3.2 days) due to influenza in
these children would translate into US ∼$300 in lost wages.
by guest on November 3, 2015
Influenza in Children • JID 2004:190 (15 October) • 1373 Download full-text
Our data can be used to help determine the direct and in-
direct costs of influenza, which are necessary for assessment of
the cost-effectiveness of influenza vaccination in children. The
results demonstrate that influenza causes a substantial burden
of illness on children and their families in terms of morbidity,
complications, treatment, and absenteeism. For each of these
aspects, the burden of illness is greatest in children !3 years
old. The findings imply that the vaccination of children !3
years old might be beneficial in reducing the total burden of
pediatric influenza on society.
We are indebted to the other members of the study group: study physi-
cians Minna Aaltonen, Matti Ahonen, Janne Kataja, Riku Kiviranta, Mik-
ko Lintu, Jussi Niemela ¨, Esa Partanen, Jaakko Pulkkinen, and Otto Rahko-
nen; study nurses Satu Heikkinen, Susanna Lehtonen, Jaana Marku, Maria
Marttila, Anne Riihima ¨ki, and Kirsi-Maija Suomela; virological researchers
Eeva Broberg, Katja Rannikko, and Aimo A. Salmi; and data managerTanja
1. Monto AS, Sullivan KM. Acute respiratory illness in the community:
frequency of illness and the agents involved. Epidemiol Infect 1993;
2. Glezen WP, Taber LH, Frank AL, Gruber WC, Piedra PA. Influenza
virus infections in infants. Pediatr Infect Dis J 1997;16:1065–8.
3. Hurwitz ES, Haber M, Chang A, et al. Studies of the 1996–1997 inac-
tivated influenza vaccine among children attending day care: immuno-
logic response, protection against infection, and clinical effectiveness. J
Infect Dis 2000;182:1218–21.
4. Glezen WP, Couch RB. Interpandemic influenza in the Houston area,
1974–76. N Engl J Med 1978;298:587–92.
5. Monto AS, Davenport FM, Napier JA, Francis T Jr. Modification of
an outbreak of influenza in Tecumseh, Michigan, by vaccination of
schoolchildren. J Infect Dis 1970;122:16–25.
6. Longini IM Jr, Koopman JS, Monto AS, Fox JP. Estimating household
7. Hurwitz ES, Haber M, Chang A, et al. Effectiveness of influenza vac-
cination of day care children in reducing influenza-related morbidity
among household contacts. JAMA 2000;284:1677–82.
8. Reichert TA, Sugaya N, Fedson DS, Glezen WP, Simonsen L, Tashiro
M. The Japanese experience with vaccinating schoolchildren against
influenza. N Engl J Med 2001;344:889–96.
9. Neuzil KM, Hohlbein C, Zhu Y. Illness among schoolchildren during
work, and secondary illness in families. Arch Pediatr Adolesc Med 2002;
10. Neuzil KM, Mellen BG, Wright PF, Mitchel EF Jr, Griffin MR. The effect
of influenza on hospitalizations, outpatient visits, and courses of anti-
biotics in children. N Engl J Med 2000;342:225–31.
11. Izurieta HS, Thompson WW, Kramarz P, et al. Influenza and the rates
of hospitalization for respiratory disease among infants and young chil-
dren. N Engl J Med 2000;342:232–9.
12. Neuzil KM, Zhu Y, Griffin MR, et al. Burden of interpandemic influ-
enza in children younger than 5 years: a 25-year prospective study. J
Infect Dis 2002;185:147–52.
13. Chiu SS, Lau YL, Chan KH, Wong WHS, Peiris JSM. Influenza-related
hospitalizations among children in Hong Kong. N Engl J Med 2002;
14. McIntosh K, Lieu T. Is it time to give influenza vaccine to healthy
infants? N Engl J Med 2000;342:275–6.
15. Centers for Disease Control and Prevention. Prevention and control
of influenza: recommendations of the Advisory Committee on Im-
munization Practices. MMWR Morb Mortal Wkly Rep 2002;51(RR-
16. Centers for Disease Control and Prevention. ACIP expands recom-
mendation for vaccinating children in the 2004–05 flu season.Available
at: http://www.cdc.gov/flu/professionals/acip/acipchild0405.htm (Ac-
cessed 8 March 2004).
17. Heikkinen T, Marttila J, Salmi AA, Ruuskanen O. Nasal swab versus
nasopharyngeal aspirate for isolation of respiratory viruses. J Clin Mi-
18. Waris M, Ziegler T, Kivivirta M, Ruuskanen O. Rapid detection of
respiratory syncytial virus and influenza A virus in cell cultures by
immunoperoxidase staining with monoclonal antibodies. J Clin Mi-
19. Simonsen L, Fukuda K, Schonberger LB, Cox NJ. The impact of in-
fluenza epidemics on hospitalizations. J Infect Dis 2000;181:831–7.
20. Wright PF, Thompson J, Karzon DT. Differing virulence of H1N1 and
H3N2 influenza strains. Am J Epidemiol 1980;112:814–9.
21. World Health Organization National Influenza Centre in Finland. In-
fluenza activity during the 2000–2001 season in Finland. Available at:
http://www.ktl.fi/flu/flu01f.htm (Accessed 9 March 2004).
22. World Health Organization National Influenza Centre in Finland. In-
fluenza activity and influenza viruses during the 2001–2002 epidemic
season in Finland. Available at: http://www.ktl.fi/flu/flu02f.htm (Ac-
cessed 9 March 2004).
serologic, or molecular detection of influenza. Arch Intern Med 2001;161:
24. Whitley RJ, Hayden FG, Reisinger KS, et al. Oral oseltamivirtreatment
of influenza in children. Pediatr Infect Dis J 2001;20:127–33.
25. Ruuskanen O, Arola M, Putto-Laurila A, et al. Acute otitis media and
respiratory virus infections. Pediatr Infect Dis J 1989;8:94–9.
26. Heikkinen T, Ruuskanen O, Waris M, Ziegler T, Arola M, Halonen P.
Influenza vaccination in the prevention of acute otitis media in chil-
dren. Am J Dis Child 1991;145:445–8.
27. Peltola V, Ziegler T, Ruuskanen O. Influenza A and B virus infections
in children. Clin Infect Dis 2003;36:299–305.
28. Gates GA. Cost-effectiveness considerations in otitis media treatment.
Otolaryngol Head Neck Surg 1996;114:525–30.
29. Niemela ¨ M, Uhari M, Mo ¨tto ¨nen M, Pokka T. Costs arising from otitis
media. Acta Paediatr 1999;88:553–6.
30. Capra AM, Lieu TA, Black SB, Shinefield HR, Martin KE, Klein JO.
Costs of otitis media in a managed care population. Pediatr Infect Dis
31. Cohen GM, Nettleman MD. Economic impact of influenzavaccination
in preschool children. Pediatrics 2000;106:973–6.
32. Szucs TD. Influenza: the role of burden-of-illness research. Pharma-
coeconomics 1999;16(Suppl 1):27–32.
33. Nichol KL. The efficacy, effectiveness and cost-effectiveness of inac-
tivated influenza virus vaccines. Vaccine 2003;21:1769–75.
by guest on November 3, 2015