Full text of DOI:10.2165/00019053-200119020-00008

Pharmacoeconomics 2001; 19 (2): 215-222
1170-7690/01/0002-0215/$22.00/0
ORIGINAL RESEARCH ARTICLE
©
Adis International Limited. All rights reserved.
Cost-Effectiveness Analysis of
Pneumococcal Vaccination for Elderly
Individuals in The Netherlands
1
2
3
Maarten J. Postma, Marie-Louise A. Heijnen and Johannes C. Jager
1
2
3
Groningen University Institute for Drug Exploration/Groningen Research Institute of Pharmacy
†
(
GUIDE/GRIP), Groningen, The Netherlands
Department of Infectious Diseases Epidemiology, National Institute of Public Health and the
Environment (Rijksinstituut voor Volksgezondheid & Milieu; RIVM), Bilthoven, The Netherlands
Department for Health Services Research, National Institute of Public Health and the
Environment, Bilthoven (RIVM), The Netherlands
Abstract
Objective: To assess the cost effectiveness (net costs per life year gained) of
pneumococcal vaccination of elderly individuals aged 65 years and over in The
Netherlands.
Design and setting: A pharmacoeconomic analysis was conducted from the health-
care perspective in The Netherlands. The gender- and age-specific modelling
framework linked epidemiological aspects of invasive pneumococcal disease (e.g.
incidence, mortality, life years lost) to vaccination and hospital resource use. To
derive 90% confidence limits for net costs per life year gained a stochastic analy-
sis was performed.
Intervention: Pneumococcal vaccination in the elderly with the 23-valent vac-
cine. Effectiveness of the vaccine in preventing invasive pneumococcal disease
was derived from international studies.
Main outcome measures and results: Pneumococcal vaccination in the elderly
was not found to be cost saving. At baseline, stochastic and univariate sensitivity
analysis net costs per life year gained were estimated to be between 6000 and 16 000
euro (EUR) [EUR1 = 1.1 US dollars; cost level 1995]. A scenario analysis on
alternative age-dependent vaccination strategies indicated even higher net costs
per life year gained, up to EUR28 000 for vaccinating only those elderly aged 85
years and over.
Conclusions: Pneumococcal vaccination is associated with net costs per life year
gained of EUR10 100 (at baseline assumptions). These costs are higher than those
for influenza vaccination (EUR5500). Our pharmacoeconomic approach, which
needs to be considered in conjunction with social, psychological and budgetary
issues, is intended to contribute to rational decision-making in healthcare policy.
†
Dr Postma performed part of this study while employed at the National Institute of Public Health and the Environment,
Bilthoven, The Netherlands.

2
16
Postma et al.
Growing interest exists for the pharmacoecono-
it could be estimated that less than 1% of those aged
65 years or more were vaccinated between 1994 to
1998 (Groningen University Institute for Drug Ex-
ploration, ‘InterActie databank’, unpublished data).
A Dutch study reported that vaccinating the el-
derly with pneumococcal vaccine was attractive from
an economic point of view.[ However, this study
assumed that the vaccine had a protective value
against all pneumococcal pneumonias, including
those without invasive complications. There is, how-
mic evaluation of infectious disease control in gen-
eral, and vaccination in the elderly as a specific part
of it.[ For example, the Dutch decision in 1996 to
extend influenza vaccination from high-risk elderly
to all elderly was partly based on a favourable cost-
effectiveness analysis estimated by means of a com-
1]
8]
[2]
puterised economic model. Current Dutch discus-
sionsarefocusingonthepossibleextensionofroutine
vaccination in the elderly with a vaccine against
pneumococcal infection (Streptococcus pneumon-
iae).
ever, no evidence for a protective effect of the vac-
cine for non-invasive infection.[
3,9,10]
Therefore, this
particular study may have overestimated the bene-
fits of vaccination.
Pneumococcal infection is very common in the
elderly, e.g. over 50 per 100 000 individuals aged
In the present study, we assessed the cost effec-
tiveness of pneumococcal vaccination of elderly
individuals; we limited the scope to investigating
the effects of pneumococcal vaccination to inva-
sive infection only.
6
5 years and over experience pneumococcal infec-
[
3]
tion per year in the US. In adults, the most prev-
alent manifestation of the infection is pneumonia.
Invasive pneumococcal infection, isolation of S.
pneumoniae from a normally sterile body site, re-
sults as a complication in 20% of pneumococcal
pneumonia cases. The most prevalent manifesta-
tions of invasive disease are bacteraemia (80 to 90%
of cases) and meningitis. On rare occasions, septic
arthritis and pleuritis may occur. The case fatality
for invasive disease is high in the elderly; indeed,
Research Design and Methods
The Model
General Design
The model for the economic evaluation of pneu-
mococcal vaccination in The Netherlands links epi-
demiological aspects of invasive pneumococcal
infection to healthcare costs and benefits of vacci-
nation. The epidemiological aspects include inci-
dence and case fatality. With regard to benefits of
vaccination, averted invasive infection, hospital in-
patient days and mortality were considered. Analysis
was performed for a cohort of elderly individuals
during a 5-year timeframe. Due to waning, immunity
after 5 years re-vaccination might be considered
necessitating another cost-effectiveness analysis; this
was not considered in this paper. All analyses were
gender- and age-specific (65 to 69, 70 to 74, 75 to
79, 80 to 84, and 85 years and over). Cost effec-
tiveness was expressed as net costs per life-year
gained (LYG) and was estimated at baseline and in
the sensitivity analysis.
[
3]
rates between 20 and 40% have been reported. In
addition, significant healthcare requirements, such
ashospitalisations, are associatedwithinvasivepneu-
mococcal disease.
To prevent invasive pneumococcal disease a 23-
valent polysaccharide vaccine is available. Vacci-
nation potentially saves life years and averts health-
[
4]
care utilisation. A cost-effectiveness analysis in
the US by Sisk et al.[5] indicated cost-saving poten-
tial for pneumococcal vaccination in the elderly by
preventing bacteraemia. In the US and several Eu-
ropean countries it is recommended that the vaccine
be administered to all individuals aged 65 years
and over.[
3,6]
The Dutch recommendation for the
[7]
use of the vaccine addresses specific groups at risk.
Current use of the 23-valent vaccine is low in The
Netherlands, although registration has existed since
1
984. Approximately 3000 doses per year were dis-
Epidemiological Considerations
Two recent Dutch studies estimated an annual
incidence of invasive pneumococcal disease of more
[
6]
tributed in 1994 to 1996. From a regional data-
base of pharmacists in the north of The Netherlands
©
Adis International Limited. All rights reserved.
Pharmacoeconomics 2001; 19 (2)

Cost-Effectiveness Analysis of Pneumococcal Vaccination
217
than 40 per 100 000 among individuals aged 65
years or over, based on public health laboratory
hospital case fatality in our model (table I); this
omitted possible out-of-hospital mortality, the po-
tential impact of which is considered in the section
on sensitivity analysis.
[
11,12]
surveillance over the period 1990 to 1998.
For
the period 1995 to 1997, this estimate amounted to
approximately 50 per 100 000 elderly. The national
Dutch registration of hospitalisations (SIG Health-
care Information, Utrecht; comprising all Dutch hos-
pitals except 2 that are specialised in the treatment
of cancer) reported an average of almost 1000 an-
nual pneumococcal-related hospitalisations over the
period 1995 to 1997 (48 cases per 100 000 elderly)
Economic Considerations
The average length of hospital stay (ALOS) for
pneumococcal-related hospitalisationsfor theelderly
was 18.9 days during 1995 to 1997 (SIG Health-
care Information, Utrecht). Gender- and age-specific
ALOS was inserted in the model (table I). For both
men and women, and for all ages, it was assumed
that a patient with invasive pneumococcal infec-
tion required 1.3 days in intensive care.[
[
SIG Healthcare Information, Utrecht, personal com-
munication]. For pneumococcal-related hospitalis-
ations, the International Classification of Diseases,
version 9 (ICD-9) codes 38.2, 320.1 and 481 were
used. As hospital data closely relate to subsequent
healthcare utilisation and costing estimations, we
used these as incidence estimates for the cohort of
elderly considered in the analysis (table I). As these
data are in line with the public health surveillance
data, application of the latter would not dramati-
cally alter our analysis.
14]
Unit costs were expressed in 1995 values. Unit
costs for a regular hospital inpatient day were esti-
mated at 270 euro (EUR) [EUR1 = $US1.1 = 2.2
Dutch guilders].[ Costs of an inpatient day in in-
15]
[
16]
tensive care were estimated at EUR1500. Both
estimates included standard pharmacotherapy, but
excluded payments for physician services. Per
person costs of vaccination comprised the general
practitioner (GP) charge for vaccine administra-
tion (EUR7.5), the vaccine market price excluding
value-added tax (EUR12) and the dispensing fee
A cohort of 2.06 million elderly, as of 1 January
996, were included in this analysis. During a 5-
1
year timeframe this cohort reduced from 2.06 mil-
lion at the start of the first year to 1.54 million at
the start of the fifth year (using standard life tables
for 1996 from the Dutch Central Bureau of Statis-
tics, Rijswijk[ ). In this demographic model, mor-
tality was assumed at the end of each year.
[16,17]
for the pharmacist (EUR5).
In our model we inserted small costs for mild
adverse effects following vaccination that would
possibly require a GP visit of EUR15 (1 per 100
13]
[18]
vaccinations).
The case fatality of invasive pneumococcal in-
Vaccine Characteristics
fection in the elderly has been estimated to reach
The efficacy of the 23-valent vaccine in pre-
venting invasive pneumococcal disease in the el-
3]
4
0%.[ The in-hospital case fatality registered by
SIG Healthcare Information was 17.3% during 1995
to 1997. We inserted gender- and age-specific in-
derly was estimated at approximately 80% for the
serotypes included in the vaccine.[
4,19,20]
Levels of
Table I. Gender- and age-specific hospitalisations per 100 000 elderly individuals, average length of stay (ALOS) and in-hospital case fatality
(
(
%) over the period 1995 to 1997 for the International Classification of Diseases, version 9 (ICD-9) codes 38.2, 320.1 and 481 taken together
SIG Healthcare Information, Utrecht, personal communication)
Age (y)
Hospitalisations (per 100 000)
ALOS (days)
men
17.9
17.0
16.9
Case fatality (%)
men
49
women
21
women
19.3
19.6
20.4
20.6
men
14.2
12.7
17.4
17.4
22.6
women
12.2
13.7
16.4
19.3
65-69
70-74
75-79
80-84
85+
64
32
82
38
111
110
42
48
21.0
19.1
19.7
32.4
©
Adis International Limited. All rights reserved.
Pharmacoeconomics 2001; 19 (2)

2
18
Postma et al.
antibodies remain elevated for at least 5 years in
elderly individuals.[ However, the vaccine is not
protective against serotypes not included. For The
Netherlands, the 23 serotypes included in the vac-
cine are responsible for approximately 80% of all
invasive pneumococcal infections.[21] For the cohort
considered and during the 5 years following vacci-
nation included in our timeframe of analysis, we
assumed an average vaccine effectiveness of 64%
distribution of the case fatality data by SIG Health-
care Information and standard Dutch life tables (data
for 1996 of the Central Bureau of Statistics[ ).
The average LYG per fatal case averted changed
for any year compared with the previous one, re-
flecting demographic changes in the cohort and the
shifts in vaccine effectiveness: from 8.9 LYG in the
first year after vaccination to 9.4 LYG in the fifth
year after vaccination. This rise was primarily caused
by the increasing relative importance of averted mor-
tality in the younger age groups, as vaccine effec-
tiveness decreased rapidly for the oldest age groups
3]
13]
(
8.0 × 8.0). Lower effectiveness was investigated
in the sensitivity analysis. Waning immunity and
declining vaccine effectiveness by age were also
[5,19]
taken into account (table II).
No gender differ-
(
almost to zero for the 85 years and over group; see
ences were assumed for vaccine effectiveness.
As a protective effect of the vaccine against non-
table II). Obviously, this phenomenon offsets the
decreasing effect on LYG of the aging of the co-
hort.
invasive pneumococcal disease has not been dem-
onstrated,[
3,9,10]
it seems reasonable to assume that
Cost effectiveness was estimated from the health-
care perspective, excluding indirect costs of pro-
duction losses in paid and unpaid work. The report-
ing of cost effectiveness was based on comparing
the intervention of vaccination with the ‘do noth-
ing’ alternative. As the starting point, the financial
costs and life years lost with invasive pneumococcal
infection were estimated in the absence of vaccina-
tion. In the next step benefits of averted hospital
care and LYG were calculated. Costs of vaccina-
tion minus benefits of averted hospital costs were
divided by LYG to render the cost-effectiveness
ratio (CER; net costs per LYG). The CER was es-
timated at baseline and in the sensitivity analysis.
Finally, different scenarios were investigated for
age-specific vaccination strategies within the el-
derly population (vaccinating only those aged 75
years or more and 85 years or more)
hospitalisation might not be fully prevented. We
merely assumed that whenever the vaccine was ef-
fective in preventing an invasive infection a less
serious non-invasive pneumococcal pneumonia was
presented, hospitalisation in the intensive care unit
(
ICU) was averted, hospitalisation in the non-ICU
was shortened, and case fatality was averted. The
extent to which non-ICU hospitalisation is short-
ened is as yet unclear and was investigated over the
full range of 0 to 100% in the sensitivity analysis
(
50% in the baseline). In line with this reasoning,
we did not assume that GP care was averted by
vaccination.
Cost-Effectiveness and Sensitivity Analysis
In addition to monetary benefits, vaccination ren-
ders health gains. Health gains considered were in-
vasive infections averted, mortality averted and LYG.
LYG were estimated based on the gender and age
The LYG were not discounted for time prefer-
[
22]
ence in the baseline analysis. Monetary benefits
were discounted at 4%.[ In the sensitivity analy-
sis, life years were discounted at the same rate as
23]
Table II. Assumed effectiveness (%) ofthe23-valentpneumococcal
vaccine by age group and year after vaccination (based on data
presented by Sisk et al.[5] and Shapiro et al.[19]
)
monetary benefits and other discount rates were
Age (y)
Year after vaccination
[24]
investigated (0 and 7%).
To enable the deriva-
1
2
3
4
5
tion of confidence intervals (CIs) for the CER, we
performed a partial stochastic analysis.[ Vaccine
efficacy and case fatality were assumed to be uni-
formly distributed over the range made up by the
65-69
70-74
75-79
80-84
85+
90
80
80
70
65
85
75
70
60
50
80
70
60
50
35
75
65
50
70
20
70
60
40
30
5
24]
[25]
percentages reported in tables I and II ±5%.
©
Adis International Limited. All rights reserved.
Pharmacoeconomics 2001; 19 (2)

Cost-Effectiveness Analysis of Pneumococcal Vaccination
219
Analyses were performed using Microsoft EXCEL
for spreadsheet design and @RISK (Palisade Cor-
poration) for Monte-Carlo simulation of CIs.
2
0 000
16 000
2 000
1
8
4
000
000
0
Results
At baseline, the expected cost effectiveness was
estimated to be EUR10 100 per LYG for vaccinating
all elderly individuals aged 65 years and over (fig. 1).
The 90% CI ranged from EUR9020 to EUR11 400
per LYG. Gross investment in vaccinating all el-
derly individuals amounted to EUR51.5 million and
the estimated monetary benefits of averted hospi-
talisation were EUR10.9 million. With net costs of
EUR40.6 million, no cost-saving potential for pneu-
mococcal vaccination of the elderly was indicated
at baseline assumptions.
Univariate sensitivity analysis is also presented
in figure 1. Discounting of LYG raised the expected
CER by 27%. Ranging the discount rate for mon-
etary benefits from 0 to 7% corresponded with mi-
nor changes in the CER of –2 to +1%. Ranging the
percentage of non-ICU inpatient days averted from
Baseline
No non-I CA Ul l non-ICU
Discount LYG
Discount Dr ai st ec o0 u%n t rate 7%
Raising case fatality
Raising effectiveness
Lowering effectiveness
Fig. 1. Expected net costs per life year gained (LYG) [in euro
(EUR); 1995 values] of pneumococcal vaccination for the el-
derly in The Netherlands in the baseline and one-way sensitivity
analysis (including 90% confidence intervals): (i) baseline [non-
discounting of LYG, discounting of monetary amounts at 4%,
50% of non-intensive care unit (ICU) inpatient days averted,
case fatality according to hospital registration at 17% on aver-
age, average vaccine effectiveness at 64%]; (ii) discounting of
LYG at 4%; (iii) discounting monetary amounts at 0%; (iv) dis-
counting monetary amounts at 7%; (v) no non-ICU inpatient
days averted; (vi)all non-ICUinpatientdays averted; (vii)raising
case fatalities by 25% of baseline values; (viii) lowering vaccine
effectiveness values by 25% of baseline figures; and (ix) raising
vaccine effectiveness values by 25% of baseline figures.
0
–
to 100% corresponded to changes in the CER of
15 to +15%. Raising case fatalities by 25% of
derly individuals aged 75 years and over required
an initial investment of EUR22 million and had an
expected CER of EUR13 500 per LYG. Correspon-
ding values for vaccinating elderly individuals
aged 85 years and over were EUR5.2 million and
EUR27 900 per LYG, respectively.
baseline figures (e.g. to correct for possible out-of-
hospital mortality) resulted in a 20% decrease in
the CER. Finally, major changes in the expected
CER from –25 to 42% corresponded with changes
in the age-specific vaccine effectiveness by ±25%,
respectively, of the baseline values.
Multivariate sensitivity analysis demonstrated
CERs ranging from EUR4580 to EUR23 700. The
former estimate resulted from baseline assumptions
with the exception of a 0% discount rate for mon-
etary benefits, 100% non-ICU inpatient days averted,
Discussion
The CER of pneumococcal vaccination of el-
derly individuals aged 65 years and over was esti-
mated at EUR9020 to EUR11 400 per undiscount-
ed LYG at baseline. Univariate sensitivity analysis
indicated CERs below EUR16 000 per LYG in all
investigated cases. However, in worst case scenarios
of the multivariate sensitivity analysis, estimated
CERs surpassed EUR16 000 per LYG. Furthermore,
limitation of the vaccination programme to elderly
individuals aged 85 years and over resulted in a
CER surpassing EUR16 000 per LYG.
2
5% increase in case-fatality, and 25% increase in
effectiveness. The latter estimate reflected baseline
assumptions with the exception of a 7% discount
rate for monetary benefits and LYG, no non-ICU
inpatient days averted and 25% reduction in effec-
tiveness.
With baseline assumptions, figure 2 presents re-
sults for the scenario analysis of different strate-
gies, each directed at vaccinating specific age groups
within the elderly population. Vaccinating only el-
Net costs below EUR16 000 per LYG may re-
flect acceptable cost effectiveness. The cut-off point
for acceptable net costs per LYG differs over time
©
Adis International Limited. All rights reserved.
Pharmacoeconomics 2001; 19 (2)

2
20
Postma et al.
4
3
2
1
0 000
0 000
0 000
0 000
0
model neglected possible effects of extended influ-
enza vaccination of the Dutch elderly population
during recent years. Obviously, the possible effects
of influenza vaccination on pneumococcal infec-
tions, its complications, and the cost effectiveness
of vaccination will be a key research topic in the
coming years.[
27]
Our assumption for case fatality at an average
of 17% was lower than that reported in studies for
other countries. One influencing factor, possibly
related to a relatively low case fatality, is the rela-
65+
75+
Age (years)
85+
tively low penicillin resistance of S. pneumoniae in
Fig. 2. Expected net costs per life year gained (LYG) [in euro
EUR); 1995 values] of pneumococcal vaccination for the el-
derly in The Netherlands in the baseline and scenario analysis
including 90% confidence intervals): (i) vaccinating individuals
aged 65 years and over at baseline; (ii) aged 75 years and over;
and (iii) aged 85 years and over.
[28]
The Netherlands.
However, higher case fatali-
(
[
14]
ties have been reported in The Netherlands.
In
the study by van Ampting et al.,[ case fatality for
invasive pneumococcal disease in the elderly was
estimated at 30%. While our case fatality was based
on national hospital registration data, van Ampting
et al.[ estimated case fatality on the basis of a
close follow-up of a selected group of patients. It
is possible that patients with invasive pneumococcal
infection and an underlying chronic disease have
the latter assigned as a primary hospitalisation di-
agnosis. Patients with a chronic disease are at an
increased risk for invasive pneumococcal disease
and have been shown to exhibit relatively high case
fatality after invasive infection.[ These patients
were included in the study by van Ampting et al.,[
whereas some such patients might be missed in the
hospital registration causing potential underestim-
ation of the benefits of vaccination in our model.
Debate still continues as to whether the evidence
14]
(
14]
and between societies depending on the societies’
preferences and abilities to pay the additional costs.
For example, a Canadian study conducted in 1992
put the cut-off point for programmes to be accepted
without question at 20 000 Canadian dollars ($Can)
per LYG.[ No such cut-off point has, as yet, been
defined for The Netherlands. The CER of the pre-
viously mentioned extension of the influenza vac-
cination programme for the elderly in 1995 was
estimated to be significantly lower than EUR16 000,
26]
12]
14]
[
2]
with EUR5500 per undiscounted LYG.
We presented results in terms of net costs per
LYG. Other ways of investigating the cost effec-
tiveness of pneumococcal vaccination in the elderly
population involve net costs per invasive infection
averted or net costs per fatal invasive infection aver-
ted. Expected net costs at baseline were EUR14 800
per invasive infection averted and EUR91 800 per
fatal invasive infection averted.
for the effectiveness of vaccination on invasive infec-
[10,29]
tion is sufficient.
This debate involves both the
effectiveness of vaccination for the non–chronically
ill elderly and for the chronically ill elderly.
[10,30]
Recently, beneficial effects of pneumococcal vac-
cination have been indicated for elderly patients
Some limitations of the current decision model
have to be considered. No indirect costs of produc-
tion losses were included. As the population con-
cerned involved the elderly, primarily production
losses in terms of unpaid work will occur. The in-
clusion of indirect costs related to unpaid work would
improve the cost-effectiveness results. Finally, our
[31,32]
with chronic lung disease.
One could consider
a vaccination strategy directed at chronically ill el-
derly patients, as was previously conducted for in-
fluenza in The Netherlands. The cost effectiveness
of such a programme would, however, be nega-
tively influenced by the fact that elderly patients
with a chronic disease may have a reduced life ex-
©
Adis International Limited. All rights reserved.
Pharmacoeconomics 2001; 19 (2)

Cost-Effectiveness Analysis of Pneumococcal Vaccination
221
pectancy.[33] These elderly individuals may have
relatively poor health, and deaths from pneumo-
coccal infection in chronically ill elderly patients
are likely to represent ‘frail’ cases. We previously
applied reductions in remaining life expectancy
for chronically ill elderly patients in our cost-
effectiveness analysis of influenza vaccination in
The Netherlands.[2]
progressed in recent years and consensus is grow-
ing concerning topics that were considered contro-
versial less than 10 years ago.[ As a result, the
credibility and comparability of pharmacoeconomic
evaluations has increased, enhancing their useful-
ness for healthcare policy-making.
36]
Conclusion
Other investigators have claimed that pneumo-
coccal vaccination in the elderly population is also
effective in preventing non-invasive disease and
We have developed a gender- and age-specific
pharmacoeconomic model to estimate the cost ef-
fectiveness of pneumococcal vaccination in elderly
individuals. Pneumococcal vaccination in the el-
derly population was not found to be cost-saving
in The Netherlands. In our baseline analysis and in
the univariate sensitivity analysis, vaccinating all
elderly individuals aged 65 years and over costs
less than EUR16 000 per LYG but costs more than
the influenza vaccination. Cost effectiveness was
worse for vaccination strategies directed only at
older age groups, e.g. vaccinating elderly individ-
uals aged 85 years and over cost EUR28 000 per
LYG.
have included this in their cost-effectiveness anal-
ysis.[
8,34]
Both studies assumed a vaccine effective-
ness of approximately 60% for both invasive and
non-invasive disease. Incidence figures used in these
[8]
[34]
analyses for The Netherlands and Belgium were
approximately 10-fold those used in our analysis and
hospitalisation figures were twice ours. Obviously,
these assumptions on incidence and hospitalisations
resulted in more favourable cost-effectiveness es-
timates in these studies than in our current study.
[
34]
For Belgium, cost savings were indicated
and
for The Netherlands, small net costs per LYG were
reported (EUR1500 per LYG; 1995 values).
Our pharmacoeconomic approach, which needs
to be considered in conjunction with social, psy-
chological and budgetary issues, is intended to con-
tribute to rational decision-making in healthcare
policy for infectious disease control.
[
8]
The results of analyses for Canada and the US
on vaccination for invasive pneumococcal disease
in elderly individuals indicated cost savings (US)
[
5,35]
or low costs per LYG (Canada).
The Canadian
study provided a range for pneumococcal vaccina-
tion in Quebec of $Can900 to $Can4600 per non-
discounted LYG (1995 values). Differences related
to several relevant assumptions in the models: an-
nual incidence of invasive disease (129 per 100 000
in the Canadian study versus <50 per 100 000 in
this study), case fatality (≥30% and over in the US
study versus <20% in this study) and per person
vaccination costs ($US12 at 1993 values in the US
study versus $US22 in this study).
Acknowledgements
The authors acknowledge the advice of J. van Ampting,
MD (Department of Internal Medicine, University Hospital
Utrecht), P.B. van den Berg PharmD (GUIDE/GRIP, Gron-
ingen), Dr G. Elzinga (RIVM), Prof. Dr J.W.M. van der Meer
(Department of Internal Medicine, University Hospital Nij-
megen), Dr A.J. de Neeling(RIVM), Dr W. vanPelt (RIVM),
M.J.J.C. Poos, MSc (RIVM), Dr M.J.W. Sprenger (Healthcare
Insurance Board, Amstelveen) and Prof. Dr B.A.M. van der
Zeijst (RIVM).
In general, interest by Dutch healthcare deci-
sion-makers for pharmacoeconomic evaluations in
infectious disease prevention is growing. This is
demonstrated by current discussions regarding vac-
cination in elderly individuals, newborn hepatitis
B vaccination programmes and screening. The me-
thodology of pharmacoeconomic evaluations has
References
1. Hannoun C, editor. The socioeconomics of influenza and influ-
enzavaccination in Europe. Pharmacoeconomics1996;9Suppl.
3
: 1-81
2
. Postma MJ, Bos JM, van Gennep M, et al. Economic evaluation
of influenza vaccination: assessment for the Netherlands. Phar-
macoeconomics 1999; 16 Suppl. 1: 1-100
3
. Centers for Disease Control. Prevention of pneumococcal dis-
ease: recommendations of the Advisory Committee on Immu-
©
Adis International Limited. All rights reserved.
Pharmacoeconomics 2001; 19 (2)

2
22
Postma et al.
nization Practices (ACIP). MMWR Morb Mort Wkly Rep
997; 46 (RR-8): 1-24
. Butler JC, Breiman RF, Campbell JF, et al. Pneumococcal poly-
saccharide vaccine efficacy: an evaluation of current recom-
mendations. JAMA 1993; 270: 1826-31
22. Parsonage M, Neuburger H. Discounting and health benefits.
Health Econ 1992; 1: 71-6
23. Dutch Ministry of Finance. Handboek financiële informatie
en administratie rijksoverheid. The Hague: Ministry of Fin-
ance, 1996
24. Gold MR, Siegel JE, Russel LB, et al., editors. Cost-effective-
ness in health and medicine. New York (NY)/Oxford: Oxford
University Press, 1996
25. Doubilet P, Begg CB, Weinstein MC, et al. Probabilistic sensi-
tivity analysis using Monte Carlo Simulation. Med Decis Mak-
ing 1985; 5: 157-77
1
4
5
6
. Sisk JE, Moskowitz AJ, Whang W, et al. Cost-effctiveness of
vaccination against pneumococcal bacteraemia among elderly
people. JAMA 1997; 16: 1333-9
. Fedson DS. Pneumococcal vaccination in the United States and
2
0 other developed countries, 1981-1996. Clin Infect Dis 1998;
6: 1117-23
2
7
8
. Health Council. Advies inzake Pneumococcen-immunisatie. The
Hague: Gezondheidsraad, 1982
26. Laupacis A, Feeny D, Detsky AS, et al. How attractive does a
new technology have to be to warrant adoption and utiliza-
tion? Tentative guidelines for using clinical and economic
evaluations. Can Med Assoc J 1992; 146: 473-81
27. Honkanen PO, Keistinen T, Miettinen L, et al. Incremental ef-
fectiveness of pneumococcal vaccine on simultaneously ad-
ministered influenza vaccine in preventing pneumonia and
pneumococcal pneumonia among persons aged 65 years and
older. Vaccine 1999; 17: 2493-2500
. Baltussen R, Ament A, Leidl RM, et al. Cost-effectiveness of
vaccination against pneumococcal pneumonia in The Nether-
lands. Eur J Pub Health 1997; 7: 153-61
9
. Örtqvist Å, Hedlund J, Burman L-Å, et al. Randomized trial of
2
3-valent pneumococcal capsular polysaccharide vaccine in
prevention of pneumonia in middle-aged and elderly people.
Lancet 1998; 351: 399-403
1
0. Hirschmann JV, Lipsky BA. The pneumococcal vaccine after
28. www.earrs.rivm.nl [Accessed 2000 Dec 19]
1
5 years of use. Arch Intern Med 1994; 154: 373-7
29. Fine MJ, Smith MA, Carson CA, et al. Efficacy of pneumococ-
cal vaccination in adults: a meta-analysis of randomized con-
trolled trials. Arch Intern Med 1994; 154: 2666-77
30. Leech JA, Gervais A, Ruben FL. Efficacy of pneumococcal
vaccine in severe chronic obstructive pulmonary disease. Can
Med Assoc J 1987; 136: 361-5
1
1. De Neeling AJ, van Pelt W, Hol C, et al. Temporary increase in
incidence of invasive infection due to Streptococcus pneu-
moniae in the Netherlands. Clin Infect Dis 1999; 29: 1579-80
2. Heijnen MLA, Dorigo-Zetsma JW, Wilbrink B, et al. Respira-
toire infecties in Nederland: schatting van de incidentie van
invasieve pneumokokken-infecties op basis van ISIS. Infec-
tieziekten Bull 1998; 9: 223-5
3. Dutch Central Bureau of Statistics [online]. Available from: URL:
www.cbs.nl [Accessed 2000 Dec 14]
4. van Ampting JMA, Bouter KP, Diepersloot RJA, et al. Pneumo-
coccal bacteraemia: incidence, outcome and predisposing fac-
tors. Eur J Int Med 1998; 9: 145-50
1
31. Nichol KL. The additive benefits of influenza and pneumococ-
cal vaccinations during influenza seasons among elderly per-
sons with chronic lung disease. Vaccine 1999; 17: S91-S93
32. Nichol KL, Baken L, Wuorenma J, et al. The health and eco-
nomic benefits associated with pneumococcal vaccination of
elderly persons with chronic lung disease. Arch Intern Med
1999; 159: 2437-42
33. Postma MJ, Jager JC, Reinders A, et al. Standards for cost-ef-
fectiveness analysis: a Dutch study on influenza vaccination.
Health Syst Sci 1997; 3-4: 413-24
34. De Graeve D, Lombaert G, Goossens H. Cost-effectiveness analy-
sis of pneumococcal vaccination in adults and elderly persons
in Belgium. Pharmacoeconomics 2000; 17: 591-601
35. Conseil d’évaluation des technologies de la santé du Québec.
Cost-effectiveness and cost-utility of a pneumococcal immu-
nization program in Québec. Montréal: CETS, 1999. Report
No.: CETS 98-4 RE
36. Drummond M, Brandt A, Luce B, et al. Standardizing method-
ologies for economic evaluation in health care; practice, prob-
lems and potential. Int J Technol Assess Health Care 1993; 9:
26-36
1
1
1
1
5. Postma MJ. Assessment of the economic impact of AIDS at
national and multi-national level. Thesis at the University of
Maastricht [The Netherlands]. Zutphen: Koninklijke Wöhr-
mann bv, 1998
6. Rutten FFH, Van Ineveld BM, Van Ommen BM, et al. Kosten-
berekening bij Gezondheidszorgonderzoek; richtlijnen voor
de praktijk. Utrecht: Jan van Arkel, 1993
1
1
7. Health Care Insurance Board. Farmacotherapeutisch kompas
1
997. Amstelveen: Ziekenfondsraad, 1996
8. Honkanen PO, Keistinen T, Kivela SL. Reactions following ad-
ministration of influenza vaccine alone or with pneumococcal
vaccine to the elderly. Arch Intern Med 1996; 156: 205-8
9. Shapiro ED, Berg AT, Austrian R, et al. The protective efficacy
of polyvalent pneumococcal polysaccharide vaccine. N Engl J
Med 1991; 325: 1454-60
1
2
2
0. Farr BM, Johnston BL, Cobb D, et al. Preventing pneumococcal
bacteraemia in patients at risk. Arch Intern Med 1995; 155:
Correspondence and offprints: Dr Maarten J. Postma, Uni-
versity ofGroningen, GUIDE/GRIP, AntoniusDeusinglaan
1, 9713 AV Groningen, The Netherlands.
2
336-40
1. Netherlands Reference Laboratory for Bacterial Meningitis. Bac-
terial meningitis in The Netherlands. Amsterdam: University
of Amsterdam, 1997
E-mail: m.postma@farm.rug.nl
©
Adis International Limited. All rights reserved.
Pharmacoeconomics 2001; 19 (2)