DOI: 10.1542/peds.2004-0778
2005;115;273-279 Pediatrics
Veenhoven, Diederick E. Grobbee, Elisabeth A. M. Sanders and Anne G. M. Schilder
Carole N. M. Brouwer, A. Rianne Maillé, Maroeska M. Rovers, Reinier H.
Recurrent Acute Otitis Media: A Randomized, Controlled Trial
Effect of Pneumococcal Vaccination on Quality of Life in Children With
This information is current as of August 10, 2006
http://www.pediatrics.org/cgi/content/full/115/2/273
located on the World Wide Web at:
The online version of this article, along with updated information and services, is
rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.
Grove Village, Illinois, 60007. Copyright © 2005 by the American Academy of Pediatrics. All
and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk
publication, it has been published continuously since 1948. PEDIATRICS is owned, published,
PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly
at Leeszaal Wilhelmina on August 10, 2006 www.pediatrics.orgDownloaded from
PEDIATRICS
威
Feb 2005
VOL. 115
NO. 2
䡠
䡠
䡠
䡠
䡠
䡠
䡠
䡠
䡠
䡠
䡠
䡠
䡠
䡠
Effect of Pneumococcal Vaccination on Quality of Life in Children With
Recurrent Acute Otitis Media: A Randomized, Controlled Trial
Carole N. M. Brouwer, MD, PhD*‡; A. Rianne Maille´, PhD‡; Maroeska M. Rovers, PhD‡;
Reinier H. Veenhoven, MD*; Diederick E. Grobbee, MD, PhD‡; Elisabeth A. M. Sanders, MD, PhD§; and
Anne G. M. Schilder, MD, PhD㛳
ABSTRACT. Background. Limited effectiveness of
current treatment strategies for recurrent acute otitis me-
dia (RAOM) and increasing antibiotic resistance have
diverted attention to prevention of AOM by vaccination.
Pneumococcal vaccination for AOM seems to have only
modest clinical efficacy. Thus far, the effects on health-
related quality of life (HRQoL) or functional health sta-
tus (FHS) have not been studied.
Objective. To assess the effect of vaccination on
HRQoL or FHS.
Methods. In a double-blind, randomized, controlled
trial, 383 children 1 to 7 years old with RAOM were
vaccinated with either heptavalent pneumococcal conju-
gate vaccine followed by pneumococcal polysaccharide
vaccine (pneumococcal group: n ⴝ 190) or with hepatitis
A or B vaccines (control group: n ⴝ 193). Parents com-
pleted validated Dutch versions of 8 HRQoL and FHS
instruments assessing generic FHS (Rand, Functional
Status Questionnaire specific, and Functional Status
Questionnaire generic), otitis media–specific FHS (OM-
6), otitis media–specific child HRQoL (Numerical Rating
Scale for Child), family functioning (Family Functioning
Questionnaire), and otitis media–specific caregiver
HRQoL (Numerical Rating Scale for Caregiver). Scores
were compared at baseline and at 14 and 26 months’
follow-up.
Results. At baseline, the average AOM incidence in
the pneumococcal and control group was 5.0 (SD: 2.8) and
4.9 (SD: 2.6) episodes per year, respectively, with 38.4%
and 36.8% having suffered from >6 episodes per year.
AOM frequency decreased 4.4 episodes per year in both
groups, with a considerable and comparable improve-
ment in HRQoL and FHS. No substantial differences in
HRQoL or FHS were found between the pneumococcal
and the control group at baseline or at 14 or 26 months’
follow-up.
Conclusion. Pneumococcal vaccination has no benefi-
cial effect compared with control vaccination on either
HRQoL or FHS in children 1 to 7 years old with RAOM.
Pediatrics 2005;115:273–279; health-related quality of life,
acute otitis media, recurrent acute otitis media, functional
health status, pneumococcal vaccination.
ABBREVIATIONS. OM, otitis media; AOM, acute otitis media;
RAOM, recurrent acute otitis media; HRQoL, health-related qual-
ity of life; FHS, functional health status; Rand, Rand general health
rating index for children; FSQ, Functional Status Questionnaire;
FFQ, Family Functioning Questionnaire; NRS Caregiver, Numer-
ical Rating Scale for Caregiver; NRS Child, Numerical Rating Scale
for Child; MANOVA, multivariate analysis of variance.
A
cute otitis media (AOM) is one of the most
common infectious diseases in childhood
1–4
and has considerable impact on daily func-
tioning and health-related quality of life (HRQoL) of
the affected child and his or her family.
5–8
Because
the benefit of both medical treatment and surgery
has proved to be limited
9–13
and with resistance
against common antibiotics still on the increase,
14–17
there is much interest in developing alternative
methods to prevent AOM. Because pneumococcus is
the most common bacterial cause of otitis media
(OM), research over the past decade has focused on
pneumococcal vaccination.
18–22
Pneumococcal conju-
gate vaccination in infancy has been shown to be
(highly) effective in preventing invasive disease.
23–25
Regarding AOM, the clinical efficacy seems modest
(6–7%). A larger effect has been found in the preven-
From the *Department of Pediatrics, Spaarne Hospital Haarlem, Haarlem,
Netherlands; ‡Julius Center for Health Sciences and Primary Care, Univer-
sity Medical Center Utrecht, Utrecht, Netherlands; and Departments of
§Pediatric Immunology and 㛳Otorhinolaryngology, Wilhelmina Children’s
Hospital/University Medical Center Utrecht, Utrecht, Netherlands.
Accepted for publication Jul 22, 2004.
doi:10.1542/peds.2004-0778
No conflict of interest declared.
Address correspondence to Anne G. M. Schilder, MD, PhD, Department of
Otorhinolaryngology, Wilhelmina Children’s Hospital/University Medical
Center Utrecht, Huispost KE 04.140.5, Postbus 85090, 3508 AB, Utrecht,
Netherlands. E-mail: a.schilder@wkz.azu.nl
PEDIATRICS (ISSN 0031 4005). Copyright © 2005 by the American Acad-
emy of Pediatrics.
PEDIATRICS Vol. 115 No. 2 February 2005 273
at Leeszaal Wilhelmina on August 10, 2006 www.pediatrics.orgDownloaded from
tion of recurrent AOM (RAOM) episodes, with up to
a 12% reduction of ⱖ4 AOM episodes per year.
23,26,27
Children at risk for RAOM are assumed to benefit
most through priming of their deficient immune re-
sponse by pneumococcal conjugate vaccination.
23,27–29
Because previous studies mainly addressed the
clinical efficacy of pneumococcal vaccination regard-
ing AOM, little is known about the effects of vacci-
nation on functional health status (FHS) and HRQoL.
Assessment of such outcome is important, especially
because RAOM may be considered a chronic illness,
and HRQoL and FHS are assumed to be particularly
relevant as outcome measures.
30–32
In 1998 we started a randomized, controlled trial
on the effects of pneumococcal versus control vacci-
nation in children 1 to 7 years old who had suffered
from RAOM. This article focuses on the effects of
pneumococcal vaccination versus control vaccination
on FHS and HRQoL.
METHODS
Patients
The current study is part of a double-blind, randomized, con-
trolled trial studying the effect of pneumococcal vaccination on
FHS and HRQoL of children with RAOM alongside its clinical
efficacy. The trial was conducted at the pediatric outpatient de-
partments of a general hospital (Spaarne Hospital, Haarlem, Neth-
erlands) and an academic hospital (University Medical Center,
Utrecht, Netherlands) from April 1998 to December 2001. Children
were referred by general practitioners, pediatricians, and otolar-
yngologists or were enrolled on the caregiver’s own initiative.
Inclusion criteria were: 1 to 7 years old and a history of RAOM,
defined as having had at least 2 physician-diagnosed episodes of
AOM in the preceding year. Exclusion criteria were immunodefi-
ciency other than IgA or IgG
2
subclass deficiency; cystic fibrosis;
immotile cilia syndrome; cleft palate; chromosomal abnormalities
such as Down syndrome; or severe adverse reaction to previous
vaccinations. Informed consent was obtained from the caregivers
of all children before participation in the trial. The medical ethics
committees of both participating hospitals approved the study
protocol.
Intervention and Follow-up
After inclusion in the trial, 383 children were assigned ran-
domly to vaccination with either a 7-valent pneumococcal conju-
gate vaccine (Prevnar) followed 6 months later by a 23-valent
polysaccharide vaccine (Pneumune) (pneumococcal vaccine
group: n ⫽ 190) or with a control vaccine (recombinant hepatitis B
vaccine, Engerix-B [AE Junior], in children 12–24 months old or
hepatitis A vaccine, Havrix [AE Junior], in children 24 –48 months
old) (control vaccine group: n ⫽ 193) (Fig 1). Randomization was
balanced over age (12–24 vs 24 –84 months old) and number of
AOM episodes in the year before enrollment (2–3 vs ⱖ4 episodes).
Figure 2 reflects the flow of the study participants. Only the 2
study nurses who vaccinated the children were informed on the
type of vaccine a child received; ie, the research physicians, par-
ents, and children were unaware of the treatment received. De-
mographic data and clinical indices of the severity of OM were
recorded at enrollment. Children were seen at the outpatient
department at 7, 14, 20, and 26 months’ follow-up. At each visit,
data on episodes of physician-diagnosed AOM (based on pre-
defined criteria) and other upper respiratory tract infections, as
well as data on medical and surgical treatment of AOM, were
collected.
33
Furthermore, otoscopy and tympanometry were per-
formed by the 2 study physicians (C.N.M.B. and R.H.V.). At
enrollment and during follow-up visits at 14 and 26 months,
parents completed questionnaires assessing general FHS (Rand
General Health Rating Index for Children [Rand] and Functional
Status Questionnaire [FSQ] generic and specific) and disease-
specific FHS (OM-6) of their child and a questionnaire addressing
family functioning related to the child’s ear infections (Family
Functioning Questionnaire [FFQ]). Global HRQoL of the child and
of the caregiver with respect to the child’s ear infections was
assessed by 2 numerical rating scales (Numerical Rating Scale for
Child [NRS Child] and Numerical Rating Scale for Caregiver [NRS
Caregiver]). Details of characteristics of these instruments (Table
1) as well as data on their reliability and validity have been
described elsewhere.
46
The instruments generally were demon-
strated to be reliable and valid.
Data Analysis
The study sample size was based on expected clinical benefit of
pneumococcal vaccinations. Based on data from previous studies
in the Netherlands, 55% of patients in the control group were
estimated to have at least 1 recurrence of AOM during the 18
months of follow-up after completion of vaccinations. In view of
the multifactorial cause of AOM and assuming a potential benefit
of vaccinations similar to that of antibiotic prophylaxis and tym-
panostomy tubes, a reduction of 25%, resulting in an AOM recur-
rence rate of 40% in the pneumococcal vaccine group, was judged
clinically relevant.
33
To detect this reduction, with
␣
(2-sided) ⫽
.05 and power 80%, 352 patients would have to be randomized.
All analyses were done on the basis of intention to treat. At
baseline, the pneumococcal group and control vaccine group were
compared for differences in clinical and demographic character-
istics.
To limit the number of comparisons, the Rand (generic ques-
tionnaire) and the OM-6 (disease-specific questionnaire) were con-
sidered as primary outcome measures, based on their face valid-
ity, reliability, and responsiveness.
5,37,38,45,46
Consequently, the
other questionnaires were considered secondary outcome mea-
sures.
Because questionnaire scores generally were skewed, Mann-
Whitney tests were used to assess differences in FHS and HRQoL
scores between the pneumococcal and control vaccine groups at
baseline and at 14 and 26 months’ follow-up.
A multivariate analysis of variance (MANOVA) was performed
to detect a treatment effect for all questionnaires combined.
47
MANOVA is an extension of the common analysis of variance to
situations in which ⱖ2 dependent variables (HRQoL and FHS
scores) are included; combining data increases the power to detect
a difference. For this analysis we modeled the scores at 14 and 26
months’ follow-up.
Fig 1. Vaccination schedules.
274 PNEUMOCOCCAL VACCINATION AND QUALITY OF LIFE IN RAOM
at Leeszaal Wilhelmina on August 10, 2006 www.pediatrics.orgDownloaded from
Finally, the variables considered as possible effect modifiers
were age at inclusion (12–24 vs 24 –84 months), number of AOM
episodes in the year before enrollment (2–3 vs ⱖ4 episodes),
number of upper respiratory tract infections other than AOM in
the preceding year (⬍6vsⱖ6 episodes), symptoms of hearing-
impairment (yes/no) or language difficulties in the preceding year
(yes/no), previous ear, nose, or throat surgery (yes/no), previous
adenoidectomy (yes/no), previous tympanostomy tube insertion
(yes/no), history of antimicrobial prophylaxis (yes/no), atopy
(yes/no), number of siblings, and educational level of the caregiv-
ers (high school or higher [yes/no]). The variables were tested by
linear regression models to find potential modifiers of effect of the
intervention (independent variables) on HRQoL or FHS outcome
(dependent variable) at 14 months’ follow-up.
Fig 2. Flowchart of trial participants. *
Reasons for loss to follow-up in the
pneumococcal vaccine group were fam-
ily circumstances (2), patient burden (too
high) (3), disappointing effect from vac-
cination (1), and for unknown reasons
(2). † In the control vaccine group, loss to
follow-up was due to family circum-
stances (1), movement to an unknown
address (1), disappointing effect from
vaccination (2), patient burden (4),
“prick fear”(2), dissatisfaction with staff
(1), and development of late-onset hypo-
gammaglobulinemia (1).
TABLE 1. Characteristics of HRQoL and FHS Instruments Used in the Study
Instruments Type No. of
Items
Scale Construct(s) Measured Application in Other Studies
Generic
Rand FHS 7 Likert General health: current health, previous
health, resistance to illness
Low birth weight children,
34,35
survivors of childhood cancer,
36
asthmatic children
37,38
FSQ generic FHS 14 Likert Age-appropriate functioning and
emotional behavior
Low birth weight children,
34,39
survivors of childhood cancer,
40
asthmatic children
38,41–43
FSQ specific FHS 14 Likert Impact of illness on functioning and
behavior
Same as FSQ generic
Disease specific
OM-6 FHS 6 Likert Physical suffering, hearing loss, speech
impairment, emotional distress,
activity limitations, caregiver
concerns
Children with RAOM,
5
children
with chronic OME
5,44,45
FFQ FHS 7 Likert Parents: sleep deprivation, change of
daily or social activities, emotional
distress; family: canceling family
plans or trips; siblings: feeling
neglected, demanding extra attention
None
NRS Child HRQoL 1 Index Global well-being of child related to
AOM episodes
Children with RAOM or chronic
OME
5
NRS Caregiver HRQoL 1 Index Global well-being of parent related to
child’s AOM episodes
None
ARTICLES 275
at Leeszaal Wilhelmina on August 10, 2006 www.pediatrics.orgDownloaded from
For all analyses the statistical package of SPSS 10.1 (SPSS Inc,
Chicago, IL) was used. Questionnaire scores were transformed
into 0-to-100 scales to enhance comparability.
RESULTS
Population Characteristics
At baseline, demographic and clinical characteris-
tics of the pneumococcal and control vaccine groups
were similar (Table 2), as were the mean baseline
scores on the measures of FHS and HRQoL (Table 3).
Clinical Efficacy of Pneumococcal Vaccination
After 14 and 26 months’ follow-up, no differences
between the pneumococcal vaccine and control vac-
cine groups were observed with respect to reduction
of AOM episodes and associated use of analgesics
or antibiotics. Furthermore, the number of children
receiving tympanostomy tubes was comparable in
both groups.
33
Efficacy of Pneumococcal Vaccination on HRQoL and
FHS
After 14 months’ follow-up, the Rand showed no
significant difference between the pneumococcal and
control vaccine groups (score: 23.5 vs 23.8; P ⫽ .45).
A small but statistically significant difference was
found on the OM-6 in favor of the control vaccine
group compared to the pneumococcal vaccine group
(score: 22.3 vs 21.3; P ⫽ .002, respectively). Subse-
quent comparison of scores on the secondary generic
and disease-specific HRQoL and FHS instruments
showed no significant differences between both in-
tervention groups. After 26 months’ follow-up,
HRQoL and FHS scores of the pneumococcal and
control vaccine groups did not differ at all (Fig 3 and
Table 3).
The MANOVA on all questionnaires combined
showed a marginal significant difference at the ex-
pense of pneumococcal vaccination at 14 months’
follow-up (P ⫽ .04 with the Hotelling-Lawley Trace
test). At 26 months’ follow-up, no association was
found between the scores on all questionnaires com-
bined and type of vaccination (P ⫽ .89).
None of the possible effect modifiers showed a
significant interaction effect at 14 or 26 months’ fol-
low-up.
Figure 3 shows considerable improvements in FHS
and HRQoL in both the pneumococcal and control
vaccine groups simultaneous with a decrease in
TABLE 2. Characteristics of Study Population at Inclusion
Pneumococcal Vaccine Group Control Vaccine Group
(n ⫽ 190) SD or 95%
Confidence Interval
(n ⫽ 193) SD or 95%
Confidence Interval
Age, mo 32.8 19.3 34.8 20.1
Male gender 62.1% (55.2–69.0) 61.7% (54.8–68.6)
Age at first AOM, mo 11.2 9.4 10.8 8.4
No. of siblings 1.05 0.8 1.11 0.9
Caregiver’s education ⱖ high school* 54.4% (47.1–61.3) 52.6% (45.6–59.6)
In the year prior to inclusion
Mean number of AOM episodes per y 5.0 2.8 4.9 2.6
ⱖ6 episodes of upper respiratory tract infections per y 38.4% (31.5–45.3) 36.8% (30.0–43.6)
Pneumonia 10.0% (5.7–14.3) 16.6% (11.4–21.8)
Hearing difficulties 36.3% (29.5–43.1) 33.2% (26.6–39.8)
Speech or language difficulties 25.3% (19.1–31.5) 19.2% (13.6–24.8)
History of
Chronic airway problems or atopy† 49.5% (42.4–56.6) 51.8% (44.8–58.8)
Adenoidectomy ⫾ tonsillectomy 47.4% (40.3–54.5) 46.4% (39.4–53.4)
Tympanostomy tubes 52.6% (45.5–59.7) 48.9% (41.8–56.0)
Antimicrobial prophylaxis 15.8% (10.6–21.0) 14.5% (9.5–19.5)
Speech therapy 7.4% (3.6–11.1) 10.4% (6.1–14.7)
* Minimum educational level was high school for at least 1 of the caregivers.
† Asthma, wheezing, hay fever, or eczema.
TABLE 3. HRQoL and FHS Scores and AOM Frequency at 0, 14, and 26 Months’ Follow-up in the Pneumococcal Versus Control
Vaccine Groups
0mo 14mo 26mo
Pnc Ctrl P Value* Pnc Ctrl P Value Pnc Ctrl P Value
Generic
Rand 20.2 20.1 .63 23.5 23.8 .45 25.0 24.3 .34
FSQ generic 73.9 73.7 .85 81.6 83.6 .10 87.2 86.1 .59
FSQ specific 80.9 79.9 .57 90.0 91.5 .16 92.9 91.3 .42
Disease specific
OM-6 17.6 17.5 .93 21.3 22.3 .002 22.1 22.2 .41
NRS Child 5.3 5.4 .94 7.9 8.2 .14 8.3 8.4 .50
FFQ 25.2 25.4 .87 31.3 31.3 .78 32.1 31.9 .81
NRS Caregiver 6.1 6.6 .20 8.3 8.3 .88 7.9 8.3 .45
AOM episodes/child-year 5.0 4.9 1.4 1.0 0.6 0.5
Pnc indicates pneumococcal vaccine group; Ctrl, control vaccine group.
* Mann-Whitney test.
276 PNEUMOCOCCAL VACCINATION AND QUALITY OF LIFE IN RAOM
at Leeszaal Wilhelmina on August 10, 2006 www.pediatrics.orgDownloaded from
