RESEARCH ARTICLE
Potential Cost-Effectiveness of RSV
Vaccination of Infants and PregnantWomen
in Turkey: An Illustration Based on Bursa
Data
Koen B. Pouwels1,2*, Sefika E. Bozdemir3, Selen Yegenoglu4, Solmaz Celebi3, E.
David McIntosh5, Serhat Unal6, Maarten J. Postma1,7,8, Mustafa Hacimustafaoglu3
1 Unit of PharmacoEpidemiology & PharmacoEconomics, Department of Pharmacy, University of
a11111 Groningen, Groningen, The Netherlands, 2 Modelling & Economics Unit, Centre for Infectious Disease
Surveillance & Control, Public Health England, London, United Kingdom, 3 Division of Pediatric Infectious
Diseases, Uludağ University Faculty of Medicine, Bursa, Turkey, 4 Department of Pharmacy Management,
Hacettepe University Faculty of Pharmacy, Ankara, Turkey, 5 Department of Medicine, Imperial College
London, London, United Kingdom, 6 Department of Infectious Diseases and Clinical Microbiology,
Hacettepe University Faculty of Medicine, Ankara, Turkey, 7 Institute of Science in Healthy Aging &
healthcaRE (SHARE), University Medical Center Groningen (UMCG), Groningen, The Netherlands,
8 Department of Epidemiology, UMCG, University of Groningen, Groningen, The Netherlands
OPENACCESS
* k.b.pouwels@gmail.com
Citation: Pouwels KB, Bozdemir SE, Yegenoglu S,
Celebi S, McIntosh ED, Unal S, et al. (2016)
Potential Cost-Effectiveness of RSV Vaccination of
Infants and Pregnant Women in Turkey: An Abstract
Illustration Based on Bursa Data. PLoS ONE 11(9):
e0163567. doi:10.1371/journal.pone.0163567
Editor: Stephania A Cormier, University of Background
Tennessee Health Science Center, UNITED STATES
Worldwide, respiratory syncytial virus (RSV) is considered to be the most important viral
Received: May 18, 2016
cause of respiratory morbidity and mortality among infants and young children. Although no
Accepted: September 10, 2016 active vaccine is available on the market yet, there are several active vaccine development
Published: September 30, 2016 programs in various stages. To assess whether one of these vaccines might be a future
© asset for national immunization programs, modeling the costs and benefits of various vacci-Copyright: 2016 Pouwels et al. This is an open
access article distributed under the terms of the nation strategies is needed.
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original Objectives
author and source are credited.
To evaluate the potential cost-effectiveness of RSV vaccination of infants and/or pregnant
Data Availability Statement: All relevant data are women in Turkey.
within the paper and its Supporting Information
files.
Methods
Funding: The authors received no specific funding
for this work. A multi-cohort static Markov model with cycles of one month was used to compare the cost-
Competing Interests: ED McIntosh is a former effectiveness of vaccinated cohorts versus non-vaccinated cohorts. The 2014 Turkish birth
employee of Novartis Vaccines, which was cohort was divided by twelve to construct twelve monthly birth cohorts of equal size
developing an RSV vaccine. Since the study, ED
(111,459 new-borns). Model input was based on clinical data from a multicenter prospec-
McIntosh has become an employee of Takeda
Vaccines. Takeda Vaccines is not developing a RSV tive study from Bursa, Turkey, combined with figures from the (inter)national literature and
vaccine. MJ Postma received grants and honoraria publicly available data from the Turkish Statistical Institute (TÜÏK). Incremental cost-
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 1 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
from GSK, Pfizer, Astellas, Astra Zeneca, Sanofi, effectiveness ratios (ICERs) were expressed in Turkish Lira (TL) per quality-adjusted life
SPMSD, Vertex, SHIRE, Amgen, Novartis, year (QALY) gained.
Novavax, and Intercept, owns 3% of shares of
Ingress Health, and is advisor to Mihajlovic Health
Economics and AscAcademics. The authors Results
declare that they have no conflict of interest in
relation to this work. There are no patents, Vaccinating infants at 2 and 4 months of age would prevent 145,802 GP visits, 8,201 hospi-
products in development, or marketed products to talizations and 48 deaths during the first year of life, corresponding to a total gain of 1650
declare. This does not alter the authors’ adherence
QALYs. The discounted ICER was estimated at 51,969 TL (26,220 US $ in 2013) per
to all the PLOS ONE policies on sharing data and
materials, as detailed online in the guide for QALY gained. Vaccinating both pregnant women and infants would prevent more cases,
authors. but was less attractive from a pure economic point of view with a discounted ICER of
61,653 TL (31,106 US $ in 2013) per QALY. Vaccinating only during pregnancy would
result in fewer cases prevented than infant vaccination and a less favorable ICER.
Conclusion
RSV vaccination of infants and/or pregnant women has the potential to be cost-effective in
Turkey. Although using relatively conservative assumptions, all evaluated strategies
remained slightly below the threshold of 3 times the GDP per capita.
Introduction
In various regions of the world, respiratory syncytial virus (RSV) is a common causative agent
of childhood acute bronchiolitis and pneumonia, with frequencies varying from 12 to 52% [1].
In hospitalized infants and children<2 years of age, RSV causes up to 50–90% of acute bron-
chiolitis and 5–40% of pneumonia cases [2], [3]. Premature infants and children<2 years of
age with underlying chronic lung disease or congenital heart disease have a high risk of hospi-
talization and sequelae [4]. A study from the United Kingdom, estimated that 14% and 3% of
hospitalizations in those<2 years of age are due to RSV and influenza virus, respectively [5].
There is also evidence that bronchiolitis cases are more severe when RSV is a causative agent,
thereby further increasing the burden. In a multicenter study conducted in Finland, of the hos-
pitalized infants with bronchiolitis<2 years of age, the length of stay for RSV bronchiolitis was
longer than for other causes and ICU admissions (10%) were higher than for other causes [6].
Globally, RSV causes nearly 34 million lower respiratory tract infection (LRTI)–acute bronchi-
olitis or pneumonia—episodes and 3.4 million hospitalizations per year in infants and children
<5 years of age [7].
In Turkey, RSV is also a common cause of LRTI among infants and children. In a prospec-
tive study, 8% of all babies<6 months developed bronchiolitis caused by RSV and half of them
(4%) were hospitalized [8]. In studies from different regions of Turkey among children hospi-
talizedwith LRTI, RSV was identified as a causative agent in 35–50% of the cases [8], [9], [10].
In the Bursa region, the annual incidence of hospitalization due to RSV+ LRTI was found to be
7.81 per 1000 infants and children<2 years of age. Compared to other age groups, the burden
was highest in children aged 0–3 months, 48% of RSV+ LRTI hospitalizations were in this age-
group [9].
Vaccination has the potential to be the most effective approach for reducing the global bur-
den of disease associated with RSV infections, and the development of a safe and effective vac-
cine may have an important role in decreasing the burden of RSV infections especially in
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 2 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
infants. However, there is no commercially available active vaccine for RSV infections. Cur-
rently, there are several active vaccine development programs at various stages [11].
Given the high burden among infants and young children, an intuitive strategy would be
vaccination of infants. Vaccination of pregnant women is another potential strategy to protect
full-term infants from RSV disease during the first fewmonths following birth, by means of
transplacental transfer of antibodies [12]. A study from Turkey found that the risk of RSV
LRTIs was high among babies having low maternal antibody levels (<20RU) [8], suggesting
that maternal vaccination may be an option to decrease the substantial burden among infants
and the very young children.
To analyze whether any one of the RSV-vaccines in development might be an asset for
national immunization programs, modeling the costs and benefits of various vaccination strat-
egies is needed [12]. This cost-effectiveness study aims to evaluate the economic burden of
RSV infection among infants and children<2 years in Turkey, and to assess the potential pub-
lic health and economic benefits of RSV vaccination. Different vaccination strategies are evalu-
ated, thereby assessing the cost-effectiveness of vaccination of pregnant women and/or young
infants and exploring the potential impact of seasonal vaccination.
Material and Methods
Study population
This cost-effectiveness analysis is based on a multicenter prospective study from Bursa (Tur-
key), combined with figures from the (inter)national literature and publicly available data from
the Turkish Statistical Institute [13].
Details about the Bursa study have been described elsewhere [9]. In brief, the study was a
multicenter prospective study, in which all infants and children<2 years of age who were hos-
pitalized for LRTI in the three largest hospitals of Bursa City Center (Uludağ University, Pedi-
atric Infectious Diseases Department, Dörtçelik Children’s Hospital, and Doruk Private Bursa
Hospital) between 1 March 2010 and 28 February 2011 (12 months) were tested for RSV posi-
tivity. Patients who had received in the previous 10 days or were receiving palivizumab prophy-
laxis or intravenous immunoglobulin at the time of admission were excluded from the study.
RSV status was determined by an experiencedphysician or a trained nurse in the first 48 hours
of hospitalization with RSV Respi-Strip (Coris Bioconcept Organization). Hospitalization costs
were obtained from the documented hospital bills after discharge.
The data from this Bursa study were used to parameterize the model regarding the incidence
of RSV-associated hospitalizations and related costs, thereby assuming that these figures are
representative for the whole of Turkey. Since patients receiving palivizumab or intravenous
immunoglobulinwere excluded from that study, and accurate data about the number of chil-
dren with high-risk conditions, e.g. congenital heart disease and/or chronic heart disease, in
Turkey are lacking, we did not model an impact of vaccination on palivizumab or immuno-
globulin use.
Model design and methodological assumptions
We constructed a static multi-cohortMarkov model with cycles of one month usingMicrosoft
Office Excel 2010 to compare the cost-effectiveness of vaccinated cohorts versus non-vacci-
nated cohorts. The 2014 Turkish birth cohort was divided by twelve to construct twelve
monthly birth cohorts of equal size (111,459 new-borns) (Turkish Statistical Institute) [13].
The cohorts were followed in monthly cycles until their 2nd birthday, and clinical and eco-
nomic effects were subsequently compared. Hence the time-horizonwas 2 years. In scenarios
where RSV-associated mortality was modelled and vaccination was assumed to reduce this
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 3 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
burden, the cycles still stopped at the 2nd birthday of the birth cohorts. However, for these sce-
narios, discounted life-time years lost were included in the calculation of the incremental cost-
effectiveness ratio.
In our primary analysis we compared the cost-effectiveness of three different vaccination
strategies to the current situation of no vaccination: 1) vaccinating infants at 2 and 4 months of
age, 2) vaccinating pregnant women and 3) vaccinating pregnant women and infants at 2 and 4
months of age. We assumed that vaccinating pregnant women would occur during the third
trimester and that protection of the child started at birth, with no additional protection to the
pregnant women given the lack of reliable data about the burden RSV among pregnant
women. These are both conservative assumptions: recent animal studies suggest that RSVmay
be vertically transmitted, after which the virus may interfere with crucial developmental pro-
cesses [14]; although poorly documented,maternal RSV infectionmay lead to clinically severe
maternal disease [15].
In our model, all individuals of the unvaccinated cohort start as susceptible to symptomatic
RSV infection requiring a general practitioner (GP) visit. Each month the cohort faces a risk of
a symptomatic RSV infection requiring a GP visit. The risk is dependent on both the age of the
cohort (in months) and the season (in calendar months). A proportion of those GP-visits will
also lead to a hospitalization, of which a proportion will die due to the RSV infection. Although
re-infectionwith RSV occurs frequently, re-infections usually have a mild character with symp-
toms of uncomplicated upper respiratory tract infection [16]. Hence, we assumed that after
RSV-related GP visits patients moved to an immune state up until their 2nd birthday. Depen-
dent on the vaccination scenario, individuals of the vaccinated cohort face a monthly probabil-
ity of being vaccinated. Dependent on the assumed vaccine effectiveness, a proportion of the
vaccinated subjects move to the immune state. Some of the subjects will become susceptible
again due to waning of the vaccine-induced immunity. Finally, individuals can die in all states
of the model due to other causes than RSV. Schematic representation of the Markov model and
the corresponding RSV-related GP visits, hospitalizations and death can be found in the sup-
plementary files (S1 Appendix).
Health-care utilization
The point estimates and probability distributions used to parameterize the model are shown in
Table 1. Data from the Bursa study were used to estimate the transition probability for hospi-
talizations [9]. The anonymized electronic data of this study were used to calculate age- and
calendar-month specific transition probabilities. To remove random fluctuation due to a lim-
ited amount of hospitalizations at higher ages, we used the actual numbers for the months 0–3,
a 3-month moving average from 4–11 months, and for the months 12–23 we used the average
amount of cases observedper month during the second year of life. We considered using age-
distribution observed in larger studies from different countries [17], [18], [19], however those
studies appeared to have a relevantly different pattern in the incidence during the crucial first 4
months of life. The incidence was adjusted for imperfect sensitivity (91%) and specificity (98%)
of the Respi-Strip that was used in the Bursa study [20], using the following formula:
observed incidence  ð1  specificityÞ
senstivity þ specificity  1
There was no data on the actual number of RSV infections not resulting in health-care utili-
zation and RSV infections resulting in a GP visit.. To obtain the number of RSV-related GP
visits, we assumed that the proportion of RSV-related GP visits resulting in RSV-related
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 4 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
Table 1. Disease and vaccine parameters used in the economic model.
Base-case Distribution References
Epidemiological parameters
RSV-hospitalizations # #
RSV-related GP visits (proportion of hosp) 17.78 Beta (94.32; 1,582.46) [21]
ICU admission (as proportion of hosp) 0.10 Beta (26;228) [9]
RSV-related mortality 0–11 m (proportion of hosp) 0.0068 Beta (99.32;1,4536.27) [9], [22], [25]
RSV-related mortality 12+ m (proportion of hosp) 0.00026 Beta (99.97;390,299.03) [9], [22], [25]
Quality of life parameters
Disutility GP-treated RSV infection 0.16 Beta (56.30;281.40) [26], [27]
Disutility hospital-treated RSV infection 0.43 Beta (117.18;154.38) [26], [27]
Duration of disutility RSV infections (days) 14 Fixed [26], [27]
Average quality of life (varying with age in years) [32], [34]
<16 y 0.88 Beta (8.41;1.15)
16–24 y 0.80 Beta (11.97;2.98)
25–34 y 0.80 Beta (11.97;2.98)
35–44 y 0.77 Beta (12.75;3.72)
45–54 y 0.75 Beta (13.35;4.50)
55–64 y 0.70 Beta (14.04;6.15)
65–74 y 0.69 Beta (14.09;6.44)
75+ y 0.64 Beta (14.11–7.86)
Vaccine parameters
Vaccination coverage 0.85 Fixed Assumed
Effectiveness vaccination pregnant women 0.6 Fixed Assumed
Effectiveness vaccination infants 2 months of age 0.6 Fixed Assumed
Effectiveness vaccination infants 4 months of age 0.75 Fixed Assumed
GP: general practitioner, hosp: hospitalization, ICU: intensive care unit, m: month, RSV: respiratory syncytial virus, y: years.
# Age- and calendar-month dependent.
doi:10.1371/journal.pone.0163567.t001
hospitalizations would be the same as in the Netherlands [21]. Therefore, we estimated that for
17.78 RSV-related GP visits there will be one RSV-related hospitalization.
Mortality
Published estimates of RSV-related mortality among children<12 months old vary substan-
tially from 0 up to 22 per 100,000 person-years [21], [22], [23], [24], [25]. We used the USmor-
tality rate as estimated by Thompson et al., in the absence of Turkey-specific data. That study
estimated the RSV-related mortality rate at 5.3 per 100,000 persons for children aged<12
months, which in our model results in a case fatality ratio for children admitted to the hospital
of 5.3/781 = 0.7%. This is very similar to the average case fatality ratio of 0.7% for children aged
<12 months admitted to the hospital with RSV-associated LRTI in industrialized countries
[25].
Utilities (QALYs)
Quality of life losses were included during the acute disease episode. Although several cost-util-
ity analyses included quality of life losses during the acute phase of RSV disease, no study so far
incorporated actual utility losses estimated specific for the acute phase of RSV disease. After
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 5 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
systematically reviewing the literature, we found only one study estimating utility weights spe-
cific for this phase of the disease [26], [27]. That study estimated disutility for 4 different states
using the time-tradeoff (TTO)methodology:1) mild RSV disease resulting in emergency room
or family doctor’s office; 2) moderate RSV disease resulting in hospitalization; 3) severe RSV
disease resulting in intensive care unit admission with oxygen support by mask-assisted breath-
ing; and 4) very severe RSV disease resulting in intensive care unit admission with oxygen sup-
port by breathing machine. Assuming the same average duration of infection of 14 days, we
based the QALY loss due to an RSV-associated GP visits on the estimate for ‘mild RSV disease’.
The QALY loss due to an RSV-associated hospitalization was based on the estimates for mod-
erate RSV and very severe RSV disease, thereby using the same risk of ICU-admission as
observed in the Bursa study (10.2%). It should be noted that these estimates are obtained by a
proxy questionnaire, i.e. adults were presented with the 4 different hypothetical scenarios in
children [27].
In scenarios where we assumed that vaccination would prevent potential RSV-associated
mortality, we also included quality of life losses due to death. Because RSV-associated mortality
may be mainly found in high-risk patients with a likely reduced baseline quality of life and life-
expectancy [28], we assumed that all RSV-associated deaths would occur among patients with
a reduced quality of life and life-expectancy. Similar to studies evaluating the cost-effectiveness
of palivizumab among high-risk infants [29], [30], we calculated the life expectancy for high-
risk children based on a study evaluating the survival of infants with CHD in the United King-
dom [31]. Based on that study we assumed that 95.3% of children with CHDwould survive to
age 16 years if they had survived to age of one year. We assumed that the survival after that age
would be the same as for the general population [13], resulting in a life-expectancyof 75.0
years at the age of one year. For comparison, the life expectancy for the general population at
this age is 77.9 years.
The assumed reduced baseline quality of life of infants that die due to RSV was based on a
study comparing quality-of-life in children with a history of preterm birth and RSV hospital-
ization compared with a control group of preterm children without a history of RSV
hospitalization [32]. That study assessed the children’s health related quality of life at the age
of 5, using the Health Utilities Index (HUI) Mark 2. The median HUI 2 multi-attribute util-
ity function was 0.88 in the children with a proven RSV infection, while the median was 0.95
in the control group. In line with our assumption that the association between RSV and
asthma is not causal, we assumed that RSV does not confer long-term sequelae. Therefore
we conservatively multiplied all baseline quality of life estimates of the general population
with 0.88 to obtain the assumed reduced quality of life in children dying due to a RSV infec-
tion. In the absence of Turkish data, similar to another recent Turkish cost-effectiveness
analysis [33], age-specific population norm quality of life estimates were based on UK data
[34].
Costs
Cost estimates (Table 2) were derived from documented hospital bills after discharge in the
Bursa study and published sources. Analyses were conducted from the societal perspective,
including productivity costs. All costs were reported in 2013 Turkish Lira (TL) (1 TL = 0.50 US
$ in 2013). Costs not available at 2013 price levels were inflated using the Turkish consumer
price index.We used the same cost for GP visits and productivity losses as in a recent study
evaluating the cost-effectiveness of rotavirus vaccination in Turkey [35]. The vaccine price was
assumed to be 60 TL.
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 6 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
Table 2. Cost parameters used in the economic model.
Base-case Distribution References
Cost of GP visit (TL) 38 Fixed [35]
Cost of RSV-hospitalization (TL) 1,992.04 Gamma (1;1,992.04) [9]
Proportion of woman working 0.15 Triangular (0.1;0.15;0.2) [35]
Productivity loss per GP visit (workdays) 1 Triangular (0;1;2) [35]
Productivity loss per hospitalization (workdays) 3 Triangular (1;3;5) [35]
Cost of lost workday (TL) 28.49 Gamma (1;28.49) [35]
Price of vaccine per dose (TL) 60 Fixed Assumed
GP: general practitioner, RSV: respiratory syncytial virus, TL: Turkish Lira.
doi:10.1371/journal.pone.0163567.t002
Outcome measures and cost-utility analysis
The main outcome measure was the incremental cost-effectiveness ratio (ICER) expressed as
costs per quality-adjusted life year (QALY) gained (also sometimes referred to as the incremen-
tal cost-utility ratio). The ICER was calculated by dividing the difference in cost between the
two alternatives by the difference in their effect. It represents the average incremental cost in
TL associated with 1 QALY gained. ICERs were calculated for the different vaccination sched-
ules compared to no vaccination. Besides ICERs, the model also tracked the number of GP vis-
its, hospitalizations, and deaths due to RSV, together with associated costs and QALY losses.
Cost and health outcomes were discounted at 3% per year.
Vaccine effectiveness
Because RSV-vaccines are still under development, assumptions had to be made regarding the
vaccine effectiveness. As explained above, we considered vaccination at three different
moments: during the third trimester of pregnancy, and at 2 and 4 months of age in infants. In
our base case scenarios we assumed vaccination coverage of 85% for all vaccinations. The initial
vaccine efficacy of vaccination during pregnancy was assumed to be 60%. The first infant dose
at 2 months of age was assumed to keep the vaccine effectiveness at 60%. The second infant
dose at 4 months of age was assumed to raise the vaccine effectiveness to 75% and provide this
level of protection until the date the vaccinated child became 2 years old.
Potential herd protection effects were not included, because of insufficient data to inform
and calibrate such a model for Turkey. Moreover, transmission dynamic modelling studies
from Kenya indicate that vaccination against RSV during pregnancy or in infants does not
likely infer a substantial herd protection effect [36], [37].
Sensitivity analyses
Univariate, multivariate, scenario, and probabilistic sensitivity analyses were performed to
explore parameter uncertainty. Probabilistic sensitivity analyses were performed to evaluate
the uncertainty of the ICERs taking into account uncertainty across all parameters
simultaneously.
Transition probabilities and disutilities were inserted as beta distributions. Cost-related
parameters were inserted as fixed values when prices were fixed (vaccination costs, GP visits)
and as gamma distribution when they were estimated from a sample (hospitalizations). For
parameters for which the uncertainty had to be obtained by expert opinion, triangular distribu-
tions were used. Outcome values were generated by running the model 5,000 times. The deci-
sion uncertainty, i.e. the probability that the vaccination strategy of interest is cost-effective for
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 7 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
different threshold of the willingness to pay per QALY gained, is subsequently presented by
cost-effectiveness acceptability curves. In addition, these simulations were used to generate
95% uncertainty intervals around the ICERs.
In the univariate analysis, all relevant parameters were varied with 25% to explore the
impact of each parameter relative to each other, while holding all other parameters fixed.
We also estimated the potential cost-effectiveness of seasonal vaccination as this may be
associated with a substantial decrease in the total vaccination costs. Given the seasonality of
RSV observed in Bursa, we considered vaccination of all children aged between 2 and 6 months
in December until February, thereby assuming a vaccine effectiveness of 60% and duration of
protection of 5 months.
Ethics Statement
Ethics committee approval was received for this study from the ethics committee of Uludağ
University Medical Faculty (number: 2010-2/34). The observational study the RSV hospitaliza-
tion rates were based on was not registered in a public database. Anonymized electronic data of
that study were used to calculate age- and calender-month specific transition probabilities.
Results
Without vaccination, RSV infection among children aged 0–2 years old were modelled to cause
343,711 GP visits, 19,334 hospitalizations and 118 deaths in a Turkish birth-cohort followed
for 2 years. The total number of discounted QALYs lost would be 5,243, of which 56% was due
to deaths and 44% due to complications with related hospitalizations and GP visits. In this
timeframe, the total discounted cost mounted up to 53,123,447 TL, of which 3% was due to
productivity losses by parents.
Table 3 summarizes the impact of the three different vaccination strategies (infants at 2 and
4 months of age; pregnant women + infants at 2 and 4 months of age; pregnant women only),
using base case assumptions.
Vaccinating infants at 2 and 4 months of age would prevent 145,802 (95% CI 123,780–
169,393) GP visits, 8,201 (95% CI 7,849–8,526) hospitalizations and 48 (95% CI 38–57) deaths
during the first year of life, corresponding to a total gain of 2,172 (95% CI 1,836–2,508) dis-
counted QALYs. Incremental costs of this 2-dose infant schedule would be approximately 112
million (95% CI 81–144 million) TL. The discounted ICER was estimated at 51,969 (95% CI
35,313–69,244) TL per QALY gained, which is below the threshold of 3x GDP per capita in
Turkey (61,821 TL) [13]. Fixing all other parameters and assuming the same effectiveness for
both infant doses, the vaccine effectiveness should be at least 61% to remain below this thresh-
old. If the vaccine price would be reduced or increased by 25%, the vaccine effectiveness should
be at least 46% or 75%, respectively.
Table 3. Impact and incremental cost-effectiveness ratios of vaccination strategies.
Vaccination GP visits Hospitalizations Deaths ICER (TL/QALY)
None 343,711 19,334 118 -
2+4 m infant 197,909 11,132 70 51,969 (95% CI 35,313–68,244)
Pregnancy 285,693 16,070 95 60,638 (95% CI 45,154–76,806)
Pregnancy + 2+4m infant 157,348 8,851 54 61,653 (95% CI 44,347–79,799)
GP: general practitioner, ICER: incremental cost-effectiveness ratio, m:month, QALY: quality adjusted life-year, TL: Turkish Lira.
doi:10.1371/journal.pone.0163567.t003
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 8 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
Vaccinating both pregnant women and infants at 2 and 4 months of age would prevent an
additional 40,560 (95% CI 33,976–47,849) GP visits, 2,282 (95% CI 2,217–2,339) hospitaliza-
tions and 16 (95% CI 13–19) deaths. However, this vaccination strategy was, although still
below the threshold of 61,821 TL, less attractive from a pure economic point of view with a dis-
counted ICER of 61,653 (95% CI 44,347–79,799) TL per QALY. If the vaccine price would be
reduced or increased by 25%, and assuming an equal effectiveness for all vaccinations, the vac-
cine effectiveness should be at least 52% or 83%, respectively.
Vaccinating only during the third trimester of pregnancy resulted in an ICER of 60,638
(95% CI 45,154–76,806) TL per QALY. Although, vaccinating only during pregnancy was asso-
ciated with much less total vaccination costs than an infant schedule with vaccinations at 2 and
4 months of age (68 vs 135 million TL), the cost-effectiveness of the infant vaccination was
more favorable because this schedule would reduce the burden of RSVmore substantially.
Obviously, when the duration of protection would be longer after vaccination during preg-
nancy, more favorable ICERs could be obtained with a pregnancy only schedule. For example,
if vaccination during pregnancy would protect infants for 5 instead of 3 months the ICER
would be 35,425 (95% CI 21,342–49,915) TL per QALY, while preventing 91,126 (95% CI
78,024–105,305) GP visits, 5,126 (95% CI 4,869–5,359) hospitalizations and 36 (95% CI 29–43)
deaths compared to no vaccination.
Sensitivity analysis
Univariate sensitivity analyses showed that when varying parameters estimates by 25% in both
ways, assumed vaccine effectiveness, RSV incidence, and the assumed vaccine price were the
major determinants of the ICER (Fig 1). Other important variables were the RSV-related
mortality, and the disutility associated with a GP visit. The disutility associated with a
Fig 1. Univariate sensitivity analysis for infant vaccination at 2 and 4 months of age. The parameters are
varied 25% in both ways (dark bars: 25% increase, light bars: 25% decrease). GP: general practitioner, ICER:
incremental cost-effectiveness ratio, RSV: respiratory syncytial virus.
doi:10.1371/journal.pone.0163567.g001
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 9 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
Fig 2. Cost-effectiveness acceptability curves. Cost-effectiveness acceptiblity curves for 2+4 months
infant vaccination (solid line), vaccination of pregnant women (dotted line) and vaccination of pregnant
women + infant vaccination (dashed line). The grey vertical line indicates the 3x GDP per capita threshold
(61,821 TL per QALY gained).
doi:10.1371/journal.pone.0163567.g002
hospitalization was much less important, which can be explained by the fact that hospitaliza-
tions were much less common than GP visits in our model.
Univariate sensitivity analysis for the other vaccination schedules showed similar sensitivity
to parameters, except that the vaccine effectiveness of pregnancy vaccination also played a role
in these scenarios. For those latter scenarios the duration of protection after vaccination during
pregnancy is a major determinant of the cost-effectiveness.
Seasonal vaccination of all children aged between 2 and 6 months betweenDecember and
February resulted in an ICER of 23,965 TL per QALY, while preventing 71,729 GP visits, 4,009
hospitalizations and 28 deaths.
Probabilistic sensitivity analysis
Fig 2 shows the cost-effectiveness acceptability curves obtained via probabilistic sensitivity
analyses. The curves show the likelihood that the vaccination strategy of interest is cost-effec-
tive at different thresholds of the willingness to pay per QALY gained.. At a threshold of 61,821
TL, 90%, 51% and 44% of Monte Carlo samples are considered cost-effective for infant vaccina-
tion only, pregnant women vaccination only and combination of both strategies, respectively.
This indicates that, when taking into account uncertainty in parameter estimates, vaccination
of infants with a vaccine that has similar properties as assumed in our base-case has a high
probability of being cost-effective.
Discussion
Our analysis shows that protection of infants and young children against RSV using an effec-
tive vaccine has the potential to be cost-effective in Turkey. Taking into account both the
potential impact on the epidemiology and the potential cost-effectiveness of different vaccina-
tion strategies, a 2-dose infant vaccination schedule alone or in combination with vaccination
of pregnant women during their third trimester seems to have most potential.
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 10 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
Vaccinating of pregnant women only becomesmuch more attractive from both an econom-
ical and epidemiological point of view if the duration of protection is sufficiently long as shown
in our sensitivity analysis. In addition, inclusion of direct protection among pregnant women
in the model may substantially increase the cost-effectiveness. Influenza vaccination of preg-
nant women has previously been estimated to be potentially cost-effective even without protec-
tion of infants [38]. However, there is a lack of reliable data about the burden of RSV among
pregnant women. Nevertheless, some recent animal studies suggest that RSVmay be vertically
transmitted, after which the virusmay interfere with crucial developmental processes of the
fetus [14] and although poorly documented,maternal RSV infectionmay lead to clinically
severe maternal disease [15].
This is the first study that evaluated the potential cost-effectiveness of RSV vaccination in a
middle-income country. We considered various vaccination strategies, including vaccination
of pregnant women and/or infants and seasonal vaccination.
Our study has several potential limitations. First, similar to previous studies [21], [39], we
decided to use a static model in the absence of reliable data to inform and calibrate such a
model. Hence, we did not include potential herd immunity or potential age-shifts. Recently,
two transmission dynamic models assessed the potential epidemiological impact of RSV vacci-
nation in Kenya [36], [37]. The availability of data on transmission patterns in Kenya allowed
those researchers to better inform and calibrate the models to actual data. Nevertheless, both
studies came to different conclusions: Poletti et al. found that annual vaccination of all primary
school children is the only alternative strategy to immunization of infants [36], while Kinyanjui
et al. concluded that immunization of young children aged 5-10m was an effectivemethod of
protection of infants against hospitalizations [37]. These contrasting results and the substantial
uncertainty around the estimates of both studies illustrate that more data is needed before a
reliable transmission dynamic model can be used for a health economic evaluation of different
vaccination strategies [40]. However, if it can be proven that the vaccine has an effect on trans-
mission of the virus, a transmission dynamic model is needed to adequately capture all benefits
from different vaccination strategies, including vaccination of the elderly [41], [42].
Second, similar to previous studies [21], [39], we only included primary RSV infections,
which may have led to an underestimation of the cost-effectiveness.However, reinfections
tend to be relatively mild.
Third, we conservatively did not include RSV-associated asthma in our model. A recent
meta-analysis suggests that RSV-infection increases the risk of asthma [43]. However, it is also
possible that RSV hospitalizations do not cause subsequent asthma, but merely identify those
patients who are predisposed to develop asthma, regardless of the severity of the RSV infection.
Moreover, studies that do find an increased risk tend to show that the association decreases
with age of the children [43]. This suggests that either the asthma-symptoms caused by RSV
hospitalization are temporarily; or that RSV hospitalizations do increase the risk of asthma
only in children which such a high propensity of developing asthma that they will inevitably
develop asthma later in life due to some other environmental factor; or that the RSV hospitali-
zations just lead to an earlier diagnosis of asthma without being a causal risk factor. A recent
randomized controlled trial found that wheezing during the first year of life could be reduced
among healthy infants by providing palivizumab treatment [44]. However, more research is
needed to prove a causal relationship betweenRSV and asthma.
Fourth, in the absence of reliable data we conservatively did not model a reduction in antibi-
otic usage for respiratory infections [41], [45], of which a large proportion is caused by RSV in
this age-group [1], [41], [46]. Given the current concerns about rises in antibiotic resistance
rates worldwide [47], this could be an important factor to take into account in future analyses
if more data becomes available.
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 11 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
We conservatively only included health-care utilization for lower respiratory tract infections
in the absence of data for other relatively common outcomes caused by RSV, such as upper
respiratory tract infections [46]. Similarly, we conservatively did not include quality of life or
productivity losses associated with symptomatic RSV infections not requiring a GP visit or
hospitalization.
Another factor that makes our estimates potentially conservative is that implementation of
vaccination would have no effect on palivizumab use among high-risk infants. Adverse effects
of vaccination were not included, because it was assumed that, given the previous troubles
while testing a RSV vaccine in children [48], a RSV vaccine will only be registered if it only
causes mild local reactions.
Finally, the incidence of RSV hospitalizations was based on the incidence in the Bursa
region during a single year. There is some evidence that the RSV incidence follows a biannual
pattern in Turkey with seasons with an early strong peak alternating with late seasons with
weak activity [49]. Hence, the start of the season and the incidencemay differ for other regions
of Turkey and different years.
Our findings are similar to studies from high-income countries that concluded that vaccina-
tion of infants against RSV has the potential to be cost-effective in the Netherlands [21] and
USA [39], or even cost-saving in Spain [50]. It would be interesting to develop and validate a
simple genericmodel to enable further cost-effectiveness evaluations in a range of settings with
little extra support, as we are currently doing in the area for influenza. This requires a new proj-
ect in cooperation with information and communication specialists.
Conclusion
RSV vaccination of infants and/or pregnant women has the potential to be cost-effective in
Turkey. Although using relatively conservative assumptions, all evaluated strategies remained
below the threshold of 3 times the GDP per capita. Our study suggests that infant vaccination
alone or in combination with vaccination of pregnant women during their third trimester
seems to have most potential. When more data become available about vaccine effectiveness
against disease and transmission, with more data about the epidemiology of RSV, a dynamic
transmission model could be developed to better capture all potential benefits of vaccination.
Supporting Information
S1 Appendix. Schematic representation of theMarkov model.
(DOCX)
Author Contributions
Conceptualization:KBP SEB SY SC EDM SUMJPMH.
Formal analysis:KBP.
Methodology:KBP EDMMJP.
Resources: SEB SY SC SUMH.
Writing – original draft:KBP.
Writing – review& editing:KBP SEB SY SC EDM SUMJPMH.
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 12 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
References
1. Hacimustafaoglu M, Celebi S, Aynaci E, Koksal N, Sinirtas M, Goral G. Evaluation of RSV frequency in
acute bronchiolitis by different methods. J Pediatr Inf. 2008; 2: 156–161.
2. Hall CB. Respiratory syncytial virus and parainfluenza virus. N Engl J Med 2001; 344: 1917–1928. doi:
10.1056/NEJM200106213442507 PMID: 11419430
3. Hall CB, McCarthy CA. Respiratory syncytial virus. In: Mandell GL, Bennett JE, Dolin R, editors. Man-
dell, Douglas, and Bennett’s Principles and practice of infectious diseases, 5th edition. Philadelphia:
Churchill Livingstone; 2000. pp. 1782–1801.
4. Welliver RC Sr, Checchia PA, Bauman JH, Fernandes AW, Mahadevia PJ, Hall CB. Fatality rates in
published reports of RSV hospitalizations among high-risk and otherwise healthy children. Curr Med
Res Opin. 2010; 26: 2175–2181. doi: 10.1185/03007995.2010.505126 PMID: 20666690
5. Ajayi-Obe EK, Coen PG, Hawrami K, Aitken C, McIntosh ED, Booy R. Influenza A and respiratory syn-
cytial virus hospital burden in young children in East London. Epidemiol Infect. 2000; 136: 1046–1058.
doi: 10.1017/S0950268807009557 PMID: 17903316
6. Jartti T, Aakula M, Mansbach JM, Piedra PA, Bergroth E, Koponen P, et al. Hospital lenght-of-stay is
associated with rhinovirus etiology of bronchiolitis. Pediatr Infect Dis J. 2014; 33:829–834. doi: 10.
1097/INF.0000000000000313 PMID: 25222302
7. Hall CB, Weinberg GA, Iwane MK, Blumkin AK, Edwards KM, Staat MA, et al. The burden of respira-
tory syncytial virus infection in young children. N Engl J Med. 2009; 360:588–598. doi: 10.1056/
NEJMoa0804877 PMID: 19196675
8. Hacimustafaoglu M, Celebi S, Aynaci E, Sinirtas M, Koksal N, Kucukerdogan A, et al. The progression
of maternal RSV antibodies in the offspring. Arch Dis Child. 2004; 89: 52–53. doi: 10.1136/adc.2002.
017780 PMID: 14709507
9. Hacimustafaoglu M, Celebi S, Elmas Bozdemir Ş, Özgür T, Özcan İ, Güray A, et al. RSV Frequency in
Children below 2 Years Hospitalized for Lower Respiratory Tract Infections. Turk J Pediatr. 2013; 55:
130–139. PMID: 24192672
10. Yilmaz G, Uzel N, Işık N, Baysal SU, Aslan S, Badur S. Viral lower respiratory tract infections in chil-
dren in İstanbul, Turkey. Pediatr Infect Dis J. 1999; 18: 173. PMID: 10048697
11. Drysdale SB, Sande CJ, Green CA, Pollard AJ. RSV vaccine use—the missing data. Expert Rev Vac-
cines. 2016; 15: 149–152. doi: 10.1586/14760584.2016.1114419 PMID: 26636902
12. Anderson LJ, Dormitzer PR, Nokes DJ, Rappuoli R, Roca A, Graham BS. Strategic priorities for respi-
ratory syncytial virus (RSV) vaccine development. Vaccine. 2013; 31 Suppl 2: B209–B215. doi: 10.
1016/j.vaccine.2012.11.106 PMID: 23598484
13. Turkish statistical Institute. Available: http://www.turkstat.gov.tr/Start.do
14. Piedimonte G, Perez MK. Alternative mechanisms for respiratory syncytial virus (RSV) infection and
persistence: could RSV be transmitted through the placenta and persist into developing fetal lungs?
Curr Opin Pharmacol. 2014; 16: 82–88. doi: 10.1016/j.coph.2014.03.008 PMID: 24810284
15. Wheeler SM, Dotters-Katz S, Heine RP, Grotegut CA, Swamy GK. Maternal effects of respiratory syn-
cytial virus infection during pregnancy. Emerg Infect Dis. 2015; 21: 1951–1955. doi: 10.3201/eid2111.
150497 PMID: 26485575
16. Henderson FW, Collier AM, Clyde WA Jr, Denny FW. Respiratory-syncytial-virus infections, reinfec-
tions and immunity. A prospective, longitudinal study in young children. N Engl J Med. 1979; 300:530–
534. doi: 10.1056/NEJM197903083001004 PMID: 763253
17. Cilla G, Sarasua A, Montes M, Arostegui N, Vicente D, Pérez-Yarza E, et al. Risk factors for hospitali-
zation due to respiratory syncytial virus infection among infants in the Basque Country, Spain. Epide-
miol Infect. 2006; 134: 506–513. doi: 10.1017/S0950268805005571 PMID: 16638164
18. Rietveld E, Vergouwe Y, Steyerberg EW, Huysman MW, de Groot R, Moll HA. Hospitalization for respi-
ratory syncytial virus infection in young children: development of a clinical prediction rule. Pediatr Infect
Dis J. 2006; 25: 201–207. doi: 10.1097/01.inf.0000202135.24485.f8 PMID: 16511380
19. Hall CB, Weinberg GA, Blumkin AK, Edwards KM, Staat MA, Schultz AF, et al. Respiratory syncytial
virus-associated hospitalizations among children less than 24 months of age. Pediatrics. 2013; 132:
e341–e348. doi: 10.1542/peds.2013-0303 PMID: 23878043
20. Wybo I, Pierard D, Stevens D, Soetens O, Lauwers S. Evaluation of the performance of RSV Respi-
Strip1 in comparison with cell culture and reverse transcriptase PCR. Meeting abstract. 19th European
Congress of Clinical Microbiology and Infectious Diseases, Helsinki, Finland. 2009; May 16.
21. Meijboom MJ, Rozenbaum MH, Benedictus A, Luytjes W, Kneyber MC, Wilschut JC, et al. Cost-effec-
tiveness of potential infant vaccination against respiratory syncytial virus infection in The Netherlands.
Vaccine. 2012; 30:4691–4700. doi: 10.1016/j.vaccine.2012.04.072 PMID: 22561315
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 13 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
22. Thompson WW, Shay DK, Weintraub E, Brammer L, Cox N, Anderson LJ, et al. Mortality associated
with influenza and respiratory syncytial virus in the United States. JAMA. 2003; 289:179–186. doi: 10.
1001/jama.289.2.179 PMID: 12517228
23. Matias G, Taylor R, Haguinet F, Schuck-Paim C, Lustig R, Shinde V. Estimates of mortality attributable
to influenza and RSV in the United States during 1997–2009 by influenza type or subtype, age, cause
of death, and risk status. Influenza Other Respir Viruses. 2014; 8: 507–515. doi: 10.1111/irv.12258
PMID: 24975705
24. Tempia S, Walaza S, Viboud C, Cohen AL, Madhi SA, Venter M, et al. Mortality associated with sea-
sonal and pandemic influenza and respiratory syncytial virus among children <5 years of age in a high
HIV prevalence setting—South Africa, 1998–2009. Clin Infect Dis. 2014; 58:1241–1249. doi: 10.1093/
cid/ciu095 PMID: 24567249
25. Nair H, Nokes DJ, Gessner BD, Dherani M, Madhi SA, Singleton RJ, et al. Global burden of acute
lower respiratory infections due to respiratory syncytial virus in young children: a systematic review
and meta-analysis. Lancet. 2010; 375:1545–1555. doi: 10.1016/S0140-6736(10)60206-1 PMID:
20399493
26. Roy LMC, Bransback N, Marra C, Carr R, Chilvers M, Lynd LD. Evaluating preferences for long term
wheeze following RSV infection using TTO and best-worst scaling. Allergy Asthma Clin Immunol.
2014; 10 Suppl 1: A64. doi: 10.1186/1710-1492-10-S1-A64
27. Roy LM. Deriving health utility weights for infants with Respiratory Syncytial Virus (RSV) (T). Univer-
sity of British Columbia. Available: https://open.library.ubc.ca/cIRcle/collections/24/items/1.0074259.
28. Byington CL, Wilkes J, Korgenski K, Sheng X. Respiratory syncytial virus-associated mortality in hospi-
talized infants and young children. Pediatrics. 2015; 135: 24–31. doi: 10.1542/peds.2014-2151 PMID:
25489019
29. Bentley A, Filipovic I, Gooch K, Büsch K. A cost-effectiveness analysis of respiratory syncytial virus
(RSV) prophylaxis in infants in the United Kingdom. Health Econ Rev 2013; 3: 18. doi: 10.1186/2191-
1991-3-18 PMID: 23919494
30. Nuijten MJ, Wittenberg W, Lebmeier M. Cost effectiveness of palivizumab for respiratory syncytial
virus prophylaxis in high-risk children: a UK analysis. Pharmacoeconomics. 2007; 25: 55–71. doi: 10.
2165/00019053-200725010-00006 PMID: 17192118
31. Wren C, O’Sullivan JJ. Survival with congenital heart disease and need for follow up in adult life. Heart.
2001; 85: 438–443. doi: 10.1136/heart.85.4.438 PMID: 11250973
32. Greenough A, Alexander J, Burgess S, Bytham J, Chetcuti PA, Hagan J, et al. Health care utilisation of
prematurely born, preschool children related to hospitalisation for RSV infection. Arch Dis Child. 2004;
89: 673–678. doi: 10.1136/adc.2003.036129 PMID: 15210503
33. Bakir M, Türel O, Topachevskyi O. Cost-effectiveness of new pneumococcal conjugate vaccines in
Turkey: a decision analytical model. BMC Health Serv Res. 2012; 12: 386. doi: 10.1186/1472-6963-
12-386 PMID: 23137037
34. Chapter 5: Self-reported health. Health survey for England 1996. Available: http://www.archive.official-
documents.co.uk/document/doh/survey96/tab5-29.htm.
35. Bakir M, Standaert B, Turel O, Bilge ZE, Postma MJ. Estimating and comparing the clinical and eco-
nomic impact of paediatric rotavirus vaccination in Turkey using a simple versus and advanced model.
Vaccine. 2013; 31: 979–986. doi: 10.1016/j.vaccine.2012.11.071 PMID: 23219433
36. Poletti P, Merler S, Ajelli M, Manfredi P, Munywoki PK, Nokes D, et al. Evaluating vaccination strate-
gies for reducing infant respiratory syncytial virus infection in low-income settings. BMC Med. 2015;
13: 49. doi: 10.1186/s12916-015-0283-x PMID: 25857701
37. Kinyanjui TM, House TA, Kiti MC, Cane PA, Nokes DJ, Medley GF. Vaccine induced herd immunity for
control of respiratory syncytial virus disease in a low-income country setting. PLOS ONE. 2015; 10:
e0138018. doi: 10.1371/journal.pone.0138018 PMID: 26390032
38. Jit M, Cromer D, Baguelin M, Stowe J, Andrews N, Miller E. The cost-effectiveness of vaccinating preg-
nant women against seasonal influenza in England and Wales. Vaccine. 2010; 29: 115–122. doi: 10.
1016/j.vaccine.2010.08.078 PMID: 21055501
39. Régnier SA. Respiratory syncytial virus immunization program for the United States: impact of perfor-
mance determinants of a theoretical vaccine. Vaccine. 2013; 31: 4347–4354. doi: 10.1016/j.vaccine.
2013.07.024 PMID: 23896421
40. Baguelin M, Flasche S, Camacho A, Demiris N, Miller E, Edmunds WJ. Assessing optimal target popu-
lations for influenza vaccination programmes: an evidence synthesis and modelling study. PLOS Med.
2013; 10: e1001527. doi: 10.1371/journal.pmed.1001527 PMID: 24115913
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 14 / 15
Potential Cost-Effectiveness of RSV Vaccination in Turkey
41. Meijboom MJ, Pouwels KB, Luytjes W, Postma MJ, Hak E. RSV vaccine in development: assessing
the potential cost-effectiveness in the Dutch elderly population. Vaccine. 2013; 31: 6254–6260. doi:
10.1016/j.vaccine.2013.10.023 PMID: 24148573
42. Gessner BD. The cost-effectiveness of a hypothetical respiratory syncytial virus vaccine in the elderly.
Vaccine. 2000; 18: 1485–1494. doi: 10.1016/S0264-410X(99)00425-9 PMID: 10618546
43. Régnier SA, Huels J. Assocation between respiratory syncytial virus hospitalizations in infants and
respiratory sequelae: systematic review and meta-analysis. Pediatr Infect Dis J. 2013; 32: 820–826.
doi: 10.1097/INF.0b013e31829061e8 PMID: 23518824
44. Blanken MO, Rovers MM, Molenaar JM, Winkler-Seinstra PL, Meijer A, Kimpen JL, et al. Respiratory
syncytial virus and recurrent wheeze in healthy preterm infants. N Engl J Med. 2013; 368: 1791–1799.
doi: 10.1056/NEJMoa1211917 PMID: 23656644
45. Assink MD, Kiewiet JP, Rozenbaum MH, Van den Berg PB, Hak E, Buskens EJ, et al. Excess drug pre-
scriptions during influenza and RSV seasons in the Netherlands: potential implications for extended
influenza vaccination. Vaccine. 2009; 27: 1119–1126. doi: 10.1016/j.vaccine.2008.11.070 PMID:
19071185
46. Vesa S, Kleemola M, Blomqvist S, Takala A, Kilpi T, Hovi T. Epidemiology of documented viral respira-
tory infections and acute otitis media in a cohort of children followed from two to twenty-four months of
age. Pediatr Infect Dis J. 2001; 20:574–581. doi: 10.1097/00006454-200106000-00006 PMID:
11419498
47. O’Neill J. Antimicrobial resistance: tackling a crisis for the health and wealth of nations. Review on anti-
microbial resistance 2014. Available: http://amr-review.org/sites/default/files/AMR%20Review%
20Paper%20-%20Tackling%20a%20crisis%20for%20the%20health%20and%20wealth%20of%
20nations_1.pdf
48. Acosta PL, Caballero MT, Polack FP. Brief history and characterization of enhanced respiratory syncy-
tial virus disease. Clin Vaccine Immunol. 2015; 23: 189–195. doi: 10.1128/CVI.00609-15 PMID:
26677198
49. Turkish Neonatal Society. The seasonal variations of respiratory syncytial virus infections in Turkey: a
2-year epidemiological study. Turk J Pediatr. 2012; 54: 216–222. PMID: 23094529
50. Acedo L, Dı́ez-Domingo J, Moraño JA, Villanueva RJ. Mathematical modelling of respiratory syncytial
virus (RSV): vaccination strategies and budget applications. Epidemiol Infect. 2010; 138: 853–860.
doi: 10.1017/S0950268809991373 PMID: 20003640
PLOS ONE | DOI:10.1371/journal.pone.0163567 September 30, 2016 15 / 15