VALUE IN HEALTH REGIONAL ISSUES 7C (2015) 67–73
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/vhri
Price Determinants of the Tendering Process for
Pharmaceuticals in the Cyprus Market
Panagiotis Petrou, MBA, PhDc*, Michael A. Talias, PhD
HealthCare Management Programme, Open University of Cyprus, Nicosia, Cyprus
AB STR A CT
Background: Tendering has been a promising approach for procuring
pharmaceuticals. Significant price reductions have been well documented by several authors. To our knowledge, there are limited data
detailing the impact of variables in the tendering process. Objectives:
In this article, our objective was to evaluate the impact of potentially
exploratory variables, which included innovation status, total value
and volume of sales of each product, health care setting administration (hospital/outpatient), patent status (branded/generic), tendering
type, and wholesale price, on price reduction in the tendering process.
Methods: Financial data of public sector sales during 2011 were
analyzed. On the basis of these data, we selected 178 medicines with
corresponding sales of €49 million, out of a total market value of €104
million. Medicines were selected according to volume, value, and
therapeutic value across all therapeutic areas. We performed a beta
regression for the assessment of impact of variables and applied the
same methodology to different subgroups. Conclusions: The generic
status of medicines is statistically significantly associated with a
higher price reduction. Tendering type by alternative, high wholesale
prices, and high volume are robust estimators for price reduction.
Innovation status does not have any effect on price reduction. Outpatient medicines reach lower prices as compared with hospital
medicines. A rather unexpected finding is the negative correlation of
high sales value with price reduction. These findings will lead to
better understanding of the tendering framework, enabling us to
further evolve its operational capacity, aiming to generate more
savings. Moreover, our study indicates areas in which a more
optimized approach is needed.
Keywords: health expenditure, innovation, pharmaceutical pricing and
reimbursement, pharmaceuticals, tendering.
Introduction
industry. Consequently, in Europe, tendering has been used
principally in procuring bulk quantities such as vaccines and
dispensable products for hospitals, which are characterized by
low differentiation. Currently, only a few countries such as
Cyprus, Malta, and Iceland apply tendering in the outpatient
sector, whereas some other countries such as The Netherlands
and Belgium implement tendering selectively [3–6].
Cyprus is heavily fragmented into public and private health
sector care sectors, an attribute that applies to the pharmaceutical market as well. Official prices of pharmaceuticals are set by
the Ministry of Health (MoH) at the wholesale level through
external reference pricing (ERP) based on a basket consisting of
one cheap, two medium, and one expensive European Union (EU)
countries. The retail price is set by adding a regressive mark-up
pharmacy profit margin to the wholesale price. Cyprus is the only
EU country without a universal coverage national health system
(NHS). As a result, private health care costs burden patients
unless they are covered by an optional private health insurance
[7]. The long anticipated NHS has been postponed for 2017
because there are significant debates regarding its structure and
operational mode [8]. The MoH provides free health care, in
Pharmaceutical expenses are rising beyond the increase in
productivity, and they are stretching health systems beyond their
viability. Currently, pharmaceutical expenses constitute the fastest growing sector of total health expenses, being just second to
personnel expenses in term of recurrent costs [1].
Several factors contribute to this increase. The pharmaceutical market is a highly heterogeneous and flawed market [2],
imputed to its domination—value wise—by products enjoying
monopolistic and oligopolistic status. Factors that determine the
market nature are new expensive products, brand loyalty, and
increased asymmetry of information due to the level of uncertainty [2].
In this context, steep price reductions stemming out of the
market exclusivity that tendering offers to successful bidders
have made tendering a necessary medicine procurement method
[3,4]. Nevertheless, this attribute is mediated by the imperative
need of a strict and detailed legal framework and the inherent
hazard of relying on one or a few suppliers for the entire market,
while at the same time sustaining the dynamic efficiency of the
Copyright & 2015, International Society for Pharmacoeconomics and
Outcomes Research (ISPOR). Published by Elsevier Inc.
* Address correspondence to: Panagiotis Petrou, HealthCare Management Programme, Open University of Cyprus, P.O. Box 12794, 2252,
Latsia, Cyprus.
E-mail: [email protected].
2212-1099$36.00 – see front matter Copyright & 2015, International Society for Pharmacoeconomics and Outcomes Research (ISPOR).
Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.vhri.2015.09.001
68
VALUE IN HEALTH REGIONAL ISSUES 7C (2015) 67–73
public health care centers, to 85% of the individuals, based on
criteria such as income, public servant employment tenure,
socioeconomic status, and disease severity. Because some of
the eligibility criteria for free public health beneficiary status
are rather biased, this has ultimately led to a grossly uneven
access to free public health care. The MoH procures all the
medicines for public health care beneficiaries through tendering.
All procured products must be registered in Cyprus, which
implies that they carry an official wholesale price as well apart
from the procurement price (the tender price). Tender outcome is
defined as the price reduction compared with the wholesale price
of the product. On average, tender prices achieved by tendering
are 95% lower for generics and around 25% lower for branded
products. (This refers to the tender price—the winning bidding
price—compared with official wholesale price) [9]. The public
sector does not apply pharmacy mark-up profit because public
pharmacists are public servants and they are remunerated by a
fixed monthly salary. Because prices to the tender process are
submitted by marketing authorization holders (MaHs), the maximum price they can submit is the official wholesale price. The
key cost drivers of Cyprus’s public sector include oncology,
neurology, and immunosuppressive agents, whereas the corresponding key cost drivers of the private sector are branded
statins, branded angiotensin receptor blockers (ARB), branded
proton pump inhibitors, and lifestyle products [10].
Despite the longstanding implementation of tendering in
Cyprus, there is no documented analysis of exploratory variables
that influence price reductions. The field of price determinants in
the tendering process has been largely overlooked mainly
because of the perception that tendering is a short-term
approach and not a long-term health policy [11]. This is imputed
to fears raised about industry sustainability, failures in some
countries such as Belgium, and overestimated savings in The
Netherlands due to the counterpoising increase in pharmacists’
fees. All these factors, along with legal issues, hindered further
dissemination of tendering [11]. Consequently, no studies have
considered the dynamics of tendering and most specifically none
of these studies so far has measured the impact of the variables
and the extent of their impact on the tendering outcome.
Recently, an Italian study analyzed the impact of tendering on
biosimilars, with regard to the number of bidders [12]. Other
studies dealt with the conceptual framework of tendering and
with its quantitative impact [9,11]. In contrast, cross-country
variations in the prices of prescription medicines have been
studied by many authors and the field of price determinants
raised considerable interest in the recent years [13,14].
The goal of this study was to examine which factors influence
the outcome of the tender, and to what extent. By identifying key
variables, we can capitalize on these, further augmenting the
efficiency of tendering.
Methods
Study Design
We identified seven exploratory variables that may influence the
outcome of tendering and are associated either with intrinsic
attributes of the product or with extrinsic attributes of the
product stemming out of its positioning and its uptake in the
market. Price is the leading characteristic of a medicine and one
of the decisive factors for its reimbursement. Pricing is a contextsensitive topic because a standoff must be achieved between
affordable prices for the payer and profitable prices that will
warrant the R&D projects of the industry, the balance between
static and dynamic efficiency [15]. It was extensively argued that
prevailing pricing schemes such as ERP may reduce short-term
expenditure, but they are blunt instruments with regard to
rewarding the true added value of a new pharmaceutical product
[16,17]. This was the rationale behind the anticipated introduction of the value-based pricing scheme, that is, the alignment of
price with the real value of the product [18]. In this direction,
innovation status was defined as another variable [19]. Cyprus
currently does not have a dedicated health technology assessment agency. Because of this gap, and to assess the impact of
innovation on price reduction through tendering, we adopted the
Amelioration du Service Medical Rendu ([ASMR] [Improvement of
Medical Benefit]) classification of Haute Autorité de Santé in
France [20,21]. In France, pharmaceutical products are assessed
by the Commission of Pharmaceutical Evaluation on the basis of
five pillars:
1. Efficacy and safety;
2. Position of the medicine in the therapeutic strategy and the
existence or absence of therapeutic alternatives;
3. Severity of the disease;
4. Type of treatment: preventive, curative, or symptomatic; and
5. Public health impact.
After the assessment is concluded, each product is awarded
an improved medical benefit level ASMR classification ranging
from I to V as follows:
1. ASMR I major improvement (new therapeutic area, reduction
in mortality);
2. ASMR II significant improvement in efficacy and/or reduction
in adverse effects;
3. ASMR III modest improvement in efficacy and/or reduction in
adverse effects;
4. ASMR IV minor improvement; and
5. ASMR V no improvement.
We adopted the French improved medical benefit level classification for the products included in our sample to assess the
impact of innovation status. Tender type is also another important variable and currently, three types of tendering are applied
in the Cyprus health setting. The first type is the “International
Nonproprietary Name (INN) sole.” This tender type asks for a
specific medicine, by its INN name. There is virtually no competition in this setting; therefore, we do not anticipate any
significant price reduction. This, however, has to be counterbalanced by law provision, which may reject the procurement of
a product should the Drug’s Committee decide that the submitted
price is high. The second type is the “INN group.” This type
applies for the procurement of several products of a specific
therapeutic category such as aromatase inhibitors and anti–
tumor necrosis factor agents and the elaboration of treatment
guidelines based on the results. This is applied in expensive and
specialized products in categories with high dropout rate and/or
low compliance, primarily for secondary care. In this tender type,
the cheapest product is set as a first-line therapy while the other
products are set in subsequent treatment lines according to
prices. The third type is the “INN alternative.” This tender type
is used in high-volume primary care products that are perceived
to be interchangeable such as statins, angiotensin-converting
enzyme inhibitors, nonsteroidal anti-inflammatory drugs, and
proton pump inhibitors (e.g., omeprazole or lansoprazole) and
asks for only one product (usually among three or more).
We also assessed the impact of value and volume of sales on
the basis of assumption that a higher volume may yield significant savings, according to current literature in the field of pricevolume agreement [22]. In addition to the above, the hospital or
outpatient administration status and the potential interrelation
with price reduction have not been studied. In Cyprus, current
69
VALUE IN HEALTH REGIONAL ISSUES 7C (2015) 67–73
Table 1 – Regression coefficients of the beta regression model for the total sample.
Covariate
Mean model (logit link) intercept
Wholesale price
Value (log)
Tendering status
Alternative
Group
Hospital outpatient
Volume (packs) log
Drug classification patent status (generic)
Beta regression
P value
Estimate
Standard error
–1.8516
0.0004
–0.2506
0.6770
0.0001
0.6625
0.006*
o0.0001*
0.0001*
0.8248
0.4967
0.4556
0.3808
0.7064
0.2047
0.2488
0.1533
0.0505
0.2149
o0.0001*
0.045*
0.0030* (outpatient)
o0.0001*
0.0001* (for generics)
Pseudo R2: 0.3891.
* Denotes statistically significant P value of o0.05.
formulary classifies products as hospital and outpatient; therefore, we deemed it necessary to assess this variable because
hospital drugs incur additional costs to the payer related to the
administration of the product to the patient. As a result, it is
interesting to investigate whether by offering lower prices the
MaH makes up these additional costs that burden the payer.
Patent status has also been extensively documented in the
literature [23–25] and has been defined as an exploratory variable.
form, strength, and pack size of a specific molecule. Among the
selected products, 101 were assigned an ASMR status [20,21]. It is
possible for a product with more than one indication to carry two
ASMR classifications, which may vary on the basis of efficacy for
each indication. Therefore, we picked the ASMR classification for
the reimbursement indication of the respective product in
Cyprus. Among the sample, 13 products were assessed as ASMR
I, 14 products as ASMR II, 24 as ASMR III, 21 as ASMR IV, and 29
products as ASMR V.
Statistical Analysis
We used a beta regression to model the percentages of price
reduction that are bounded in the range (0%–100%). Conventional
regression analysis based on the normality assumption cannot be
used for modeling percentages; rather, their modeling can be
done more appropriately by using beta distribution regression. A
more detailed justification of beta regression properties is well
documented in the literature [26]. In our regression model, the
dependent variable is the percentage price reduction in the
tendering process, which is assumed to follow a beta distribution.
The mean parameter of the beta distribution represents the
average achieved reduction value, which was found to be
dependent on the wholesale price, the logarithm of value, the
tendering status, the hospital/outpatient, the logarithms of
packs, and the drug classification. Therefore, the regression
equation takes on the following form: Price reduction ¼ constant
þ a innovation þ b (hospital/outpatient) þ c patent status þ
d wholesale price þ e volume þ f value þ g tender type
þ error where a, b, c, d, e, f, and g are the regression coefficients.
We performed the above regression for the following five
subsets of our data set, namely:
1.
2.
3.
4.
5.
For the entire market;
For products that carry an ASMR classification;
For branded products only;
For generic products only; and
For high-value medicines (products with sales of >€1,000,000
yearly value of sales).
Data Management
We collected official sales data from the MoH during 2011, and we
selected 178 products with sales of €49 million, corresponding to
approximately 50% of annual public pharmaceutical expenditure.
Sampling was done through a stratified sampling procedure based on several stratums such as volume, value, and
therapeutic importance. Medicines refer to a specific dosage
Ethical Approval
The topic of this study did not require approval by the Bioethics
and Ethics Committee because it did not involve human subjects
or data that could lead to identification of any patient.
Results
Tables 1 to 5 list the results of the regression. Generic status had
a strong impact on price reduction, which was consistent across
all subcategories as well. Tendering by alternative (procuring only
one out of several competitive products) is associated with
statistically significant price reductions in all categories. Tendering by group demonstrated a statistically significant correlation
with price reduction only in the total sample, whereas tendering
by INN (monopoly products) did not correlate with significant
price reduction.
One consistent finding across all market segments is the
noncorrelation of innovation status with price reduction. Higher
wholesale prices are statistically significantly associated with
higher price reduction, while a notable finding is the negative
correlation of total sales value with price reduction, except in the
cohort with ASMR-assigned status. This conflicts with another
consistent finding across categories, namely, significant correlation of volume with price reduction. In the total sample, outpatient medicines were associated with a statistically significant
impact on price reduction. This also occurred in the generics-only
category. Levels of pseudo R2 were satisfactory in all analyses,
indicating that the model fit the data well.
Discussion
The findings of the article are related to other findings in the
literature [9,11,12] that acknowledge that the tendering process
offers significant price reductions. Our effort to decompose the
70
VALUE IN HEALTH REGIONAL ISSUES 7C (2015) 67–73
Table 2 – Regression coefficients of the beta regression model for branded products only.
Covariate
Mean model (logit link) intercept
Wholesale price
Value (log)
Innovation status
ASMR II
ASMR III
ASMR IV
ASMR V
Tendering status
Alternative
Group
Hospital outpatient
Volume in packs (log)
Beta regression
P value
Estimate
Standard error
–2.089
0.00028
–0.165
0.886
0.0001
0.080
0.018*
0.01*
0.040*
0.535
0.077
0.345
0.243
0.659
0.445
0.500
0.585
0.41
0.86
0.48
0.67
0.873
0.467
–0.093
0.3339
0.236
0.270
0.201
0.062
0.0002*
0.088
0.64
7.75 10–08*
Pseudo R2: 0.1848.
ASMR, Amelioration du Service Medical Rendu.
* Denotes statistically significant P value of o0.05.
integral elements of tendering led to significant findings espousing tendering dynamics and input of several variables.
Generic status (generic vs. branded) is the most important
factor, and many authors demonstrated the beneficial input of
generic penetration in health systems worldwide [11,27–29].
Lower barriers of generic entry into the market and ready
availability of many competitors attest to this finding [30]. Therefore, the entry of generics leads to steep price reductions and it
constitutes a significant price determinant in the context of
tendering. In particular, Kanavos and Vandoros [13] also reported
that off-patent originator brands account for a significant variation in prices. A major determinant of branded product is the
price composition because generic products are significantly
cheaper as they do not incur any research and development
expenses. This finding is also intertwined with another predictive factor, tendering by alternative (either–or), which is
undisputedly the most efficient form of tendering, although in
one case tendering by group (preferential policy) offers statistically significant price reductions as well. In general, we concluded that price reduction correlates with the level of
competition. This was also reported by Curto et al. [12] who
specified that participation of each additional bidder lowered
prices by 10%, a feature that is predominantly observed with
generics and biosimilars. This calls for more transparency in
the call for tender to facilitate the participation of more bidders.
The resulting competition or even the threat of entry of new
bidders in the process will ultimately augment the purchasing
power of the payer.
High sales volume is associated with significant price reduction, which is in line with common practices of other countries
such as price-volume agreements [21]. Presumably, an MaH is
willing to offer lower prices for bigger quantities because the
Table 3 – Regression coefficients of the beta regression model for products with ASMR classification.
Covariate
Mean model (logit link) intercept
Wholesale price
Value (log)
Innovation status
ASMR II
ASMR III
ASMR IV
ASMR V
Tendering status
Alternative
Group
Hospital outpatient
Volume in packs (log)
Drug classification patent status (generic)
Pseudo R2: 0.2497.
ASMR, Amelioration du Service Medical Rendu.
* Denotes statistically significant P value of o0.05.
Beta regression
Estimate
Standard error
P value
–7.9
3.3
–2.627 10–02
1.014
2.496 10–01
0.96062
0.43
0.000570*
0.311
3.188
–1.28
–2.8 10–01
–3.43 10–01
3.409
3.444
3.297
7.819
5.841 10–01
1.276 10–01
2.488 10–01
2.086 10–01
–0.1116
2.110 10–01
1.613 1000
0.1519
1.418 10–06
3.367 10–01
10–01
10–01
10–01
10–02
0.39
0.70
0.40
0.29
0.019253*
0.608116
0.4628
0.000496*
1.67 10–06*
71
VALUE IN HEALTH REGIONAL ISSUES 7C (2015) 67–73
Table 4 – Regression coefficients of the beta regression model for generic products only.
Covariate
Mean model (logit link) intercept
Wholesale price
Value (log)
Tendering status
Alternative
Group
Hospital outpatient
Volume in packs (log)
Beta regression
Estimate
Standard error
P value
–1.29
0.003
–0.431
0.873
0.001
0.115
0.1381
0.028*
0.0001*
0.611
0.593
1.246
0.538
0.397
0.586
0.225
0.087
0.124
0.310
3.21 10–08*
8.67 10–10*
Pseudo R2: 0.4735.
* Denotes statistically significant P value of o0.05.
payer is legally bound to procure the predefined quantity, or
sanctions apply to the payer, as in the case of Cyprus.
However, the total value of sales has a negative statistically
significant association with price reduction. One possible explanation is that high-value products achieve so through the development of a strong brand loyalty plan [31] that interacts with
patients and prescribers at the emotional level apart from the
functional benefits [32]. It was shown that advertising of pharmaceuticals comprises a strong product differentiation marketing tool that systematically lowers price sensitivity by enhancing
brand loyalty [33,34], while concomitantly increasing sales [35].
Moreover, in most of the incurable conditions, such as those
related to oncology, neurology, and rheumatology, patients are
treated with each approved pharmaceutical product sequentially
or in combination, which creates a “virtual monopoly” [36]. This
will enable the MaH to retain high prices for blockbusters. Only
for products that carry ASMR classification (101 products;
Table 3), value is not statistically associated with price reduction.
A critical question to be raised concerns how we effectively
address this “anomaly” of the tendering process. A possible
approach would be regulation of the currently unregulated
pharmaceutical marketing framework, which could resist against
aggressive promotion of pharmaceuticals, an acknowledged prerequisite of strong brand loyalty [37].
In addition to the above, high wholesale prices have a positive
relation with price reduction, suggesting that expensive products
lead to a greater price reduction. In conjunction with previous
findings, the negative correlation of wholesale value with price
reduction, we can conclude that expensive new products entering the market offer lower prices, but as their market value
increases, the price reduction diminishes. A possible explanation
is the submission of a low price to enter the market. After the
product has reached the “star”marketing phase, industry is less
keen to reduce the price, probably due to the creation of a strong
loyalty brand and rapport with patients and stakeholders (the
star marketing phase describes the high growth and profit phase
of pharmaceuticals). This finding can also be ascribed to the fact
that prices of some products are high due to small target group,
their high uncertainty, and not due to actual production costs
incurred by the company. Therefore, it seems that companies
may have the capacity to submit lower prices. It is also possible
that MaHs can sustain high prices when they take into account
the future demand for a subsequent price reduction.
Moreover, social solidarity principles often persuade people to
pay substantially more for serious health conditions, even for
marginal health gains. Pricing must also be interpreted in the
notion that ERP is considered to be a cost-containment tool and
lacks any theoretical basis. Usually other factors protrude into
Table 5 – Regression coefficients of the beta regression model for products with sales value of >€1,000,000
annually.
Covariate
Mean model (logit link) intercept
Wholesale price
Value (log)
Innovation status
ASMR II
ASMR III
ASMR IV
ASMR V
Tendering status
Alternative
Group
Hospital outpatient
Volume in packs (log)
Pseudo R2: 0.53.
ASMR, Amelioration du Service Medical Rendu.
* Denotes statistically significant P value of o0.05.
Beta regression
Estimate
Standard error
P value
–2.1095
2.3932
–2.4448
1.0865
0.1889
0.1806
0.052
o2 10–16*
o2 10–16*
0.65
0.57
0.26
0.47
0.659
0.51
0.58
0.85
0.52
0.73
0.55
0.17
0.1300
0.2934
–0.1116
2.4719
0.1852
0.1765
0.1519
0.1665
0.4828
0.0964*
0.4628
o2 10–16*
72
VALUE IN HEALTH REGIONAL ISSUES 7C (2015) 67–73
the selection of referencing countries, such as geographical
proximity and access to prices. In the World Health Organization/Health Action International (HAI) Project Medicine Prices and
Availability [38], authors stated that it is doubtful whether the
ERP is “appropriate, efficient or optimal in accordance with any
objective criterion.” The risk of dissemination of flawed pricing
also lurks, especially for countries that follow referencing countries, while industry actively pursues higher prices in first-tolaunch countries, and this creates a spillover effect. However,
low-price countries face the risk of a late launch, which often
results in loss of utility. Finally, the referencing system levels off
easily and stabilizes [39].
One of the most important findings is the lack of a correlation
between the innovation level and the price reduction. This leads
to loss of optimality because the payer may acquire less utility for
the same monetary value. The current prevailing pricing scheme
ERP has been criticized as impeding R&D because it does not
provide incentives for innovation [40]. These findings are contradictory in terms of the current shift of leading health agencies,
such as the National Health Service and the Institute for Quality
and Efficiency in Healthcare (IQWIG) [41,42], which are gradually
fostering value-based pricing, a pricing system that aligns the
price of a product to its therapeutic value, aiming to reward and
promote genuine innovation. Although value-based pricing
stalled in the United Kingdom, the underlying reasons for its
implementation are still imperative and the need for aligning the
price to the clinical value of the product is still unmet. In this
sense, we expected that products with a low innovation status
(i.e., modest improvement) would be willing to submit lower
prices given that their demand is correspondingly low; nevertheless, this did not materialize. We also expected that truly
innovative products could retain high prices, because of lack
of substitution, which was not confirmed as well. This can be
ascribed to the flawed selection of comparative products for the
tender. Therefore, it is crucial that competitive products are
considered on grounds of evidence-based medicines.
We also assessed the hospital/outpatient status of medicine
and its ability to influence the price of products. Doctors are
inclined or even pushed to prescribe medicines in Cyprus, and
overuse of prescribed medicines, such as antibiotics, has been
well documented [43]. Therefore, we anticipated that the demand
for medicines is more likely to be influenced by patient request
regarding prescription medicine, compared with hospitaladministered medicines. Hospital medicines carry a higher possibility of substitution, and we expected that hospital medicines
would display higher price elasticity. Outpatient medicines
reached a statistically significant lower level of price reduction.
This could be attributed to interaction with patients in the
prescribing process and the higher number of alternative options,
which leads to enhanced competition.
In the category with products with sales of more than
€1,000,000 each (products that are considered blockbusters in
the Cyprus context), we verified our previous findings, confirming
that the level of innovation is not a prognostic factor and is not
associated with price reduction (Table 5). This category emerges
as a vital one because it consists of key cost drivers and highly
specialized products and it demonstrates high annual increase
rate [44]. Moreover, their cost is prohibitive for out-of-pocket
procurement and patients exert significant pressure on health
agencies for reimbursement of these products. Wholesale price
still has a statistically significant association with price reduction; nevertheless, total value displays a negative statistically
significant association with price reduction, in line with previous
findings. Volume is also a strong prognostic factor.
Tendering does not interfere with the inextricable European
reference pricing scheme, which results in significant interrelation of medicines’ prices among EU countries. This constitutes a
motive for the company to consent to a reduced but locally
constrained tender price—as an outcome of a procurement
scheme—which in return avoids a spillover effect on pharmaceutical prices in other EU countries should the payer decide to
apply price reduction.
Tendering is a potent system and excels in primary care,
whereas it seems to lack sensitivity in high-value products. It
neither addresses uncertainty nor rewards innovation. The principal question is how to enrich the potency of tendering with
selectivity. A promising approach is to enrich the tendering types
and include a compulsory payback for products that do not
possess superior performance. Alternative, risk-sharing projects
[45] or other managed entry agreement schemes [46,47] may
complement the tendering system. One comparative advantage
of tendering versus risk sharing, however, is that tendering
addresses affordability, while risk sharing does not.
Limitations
We acknowledge that several other factors that have not been
considered in our current models may influence price reduction
such as fiscal pressures on the company, significance of each
product within a company’s portfolio, and colloquial pressures by
patients’ associations and media. Moreover, we adopted the
health technology assessment from France with all the inherent
risk it comprises [48].
Conclusions
Our analysis indicates that tendering is a potent system and it is
influenced by certain parameters. The impact of tendering must
be safeguarded and must be complemented with other
approaches to reach even higher levels of performance. The
insensitivity to innovation level has to be scrutinized as well as
the negative correlation of total value with price reduction.
Capitalizing on correlating factors will further enhance tendering.
The present data refer to a small country that has gained
significant expertise in tendering and succeeded in maximizing
its efficiency. Therefore, we expect that other countries, by
capitalizing on our findings, can attain significant gains.
R EF E R EN C ES
[1] OECDiLibrary. OECD health statistics. Health expenditure and
financing: health expenditure indicators. 2014. Available from: http://
www.oecd-ilibrary.org/social-issues-migration-health/data/
oecd-health-statistics/system-of-health-accounts-healthexpenditure-by-function_data-00349-en. [Accessed August 25, 2015].
[2] Thomson S, Mossialos E. Influencing demand for drugs through cost
sharing. In: Mossialos E, Mrazek M, Walley T, eds., Regulating
Pharmaceuticals in Europe: Striving for Efficiency, Equity and Quality.
England: Open University Press, England, 2004.
[3] Dylst P, Vulto A, Simoens S. Tendering for outpatient prescription
pharmaceuticals: what can be learned from current practices in
Europe? Health Policy 2011;101:146–52.
[4] Clark M, Heltzer N, Rankin J. Managing the tender process. In: Martha
Embrey, ed. MDS-3: Managing Access to Medicines and Health
Technologies. Ed. Martha Embrey Arlington, VA: Management Sciences
for Health, 2012.
[5] Rankin J, Quick JD, Muziki S, et al. Operational Principles for Good
Pharmaceutical Procurement. Geneva: World Health Organization,
1999: Essential Drugs and Medicines Policy Interagency Pharmaceutical
Coordination Group, 1999.
[6] Vogler S, Habl C. Tendering of Pharmaceuticals in EU Member States and
EEA Countries: Results from the Country Survey. 2008. Austria: OBIG.
[7] Petrou P, Talias M. A framework for applying health technology
assessment in Cyprus: thoughts, success stories, and
recommendations. Value Health Region 2013;2:273–8.
[8] Petrou P. Financial crisis as a reform mediator in Cyprus’s health
services. Eurohealth incorporating Euro Observer 2014;20:35–8.
VALUE IN HEALTH REGIONAL ISSUES 7C (2015) 67–73
[9] Petrou P, Talias M. Tendering for pharmaceuticals as a reimbursement
tool in Cyprus public health sector. Health Policy Technol
2014;3:167–75.
[10] Petrou P. The power of r – pharmaceutical sales decomposition in
Cyprus public health care sector and determinants of drug expenditure
evolution: any lessons learned? Expert Rev Pharmacoecon Outcomes
Res 2014;14:289–300.
[11] Kanavos P, Seeley L, Sotiri Vandoros S. Tender Systems for Outpatient
Pharmaceuticals in the European Union: Evidence from the Netherlands,
Germany and Belgium. London: London School of Economics, 2009.
[12] Curto S, Ghislandi S, van de Vooren K, et al. Regional tenders on
biosimilars in Italy: an empirical analysis of awarded prices. Health
Policy 2014;116:182–7.
[13] Kanavos PG, Vandoros S. Determinants of branded prescription
medicine prices in OECD countries. Health Econ Policy Law
2011;6:337–67.
[14] Leopold C, Mantel-Teeuwisse A, Seyfang L, et al. Impact of external
price referencing on medicine prices—a price comparison among 14
European countries. Southern Med Rev 2012;5:34–41.
[15] McGuire A, Raikou M, Kanavos P. Pricing pharmaceuticals: value based
pricing in what sense? Eurohealth 2008;14:3–5.
[16] Bardey D, Bommier A, Jullien B. Retail price regulation and innovation:
reference pricing in the pharmaceutical industry. J Health Econ
2010;29:303–16.
[17] Friederszick H, Tosini N, de Véricourt F, et al. An Economic Assessment
of the Relationship between Price Regulation and Incentives to
Innovate in the Pharmaceutical Industry. European School
of Management and Technology GmbH, Berlin, Germany,
ESMT White Paper.
[18] Medicines, Pharmacy & Industry Group, Department of Health. A new
value-based approach to the pricing of branded medicines. 2011.
Available from: http://webarchive.nationalarchives.gov.uk/
20130107105354/http://www.dh.gov.uk/prod_consum_dh/groups/
dh_digitalassets/@dh/@en/documents/digitalasset/dh_122793.pdf.
[Accessed April 17, 2014].
[19] Garattini L, Cornago D, De Compadri P. Pricing and reimbursement of
in-patent drugs in seven European countries: a comparative analysis.
Health Policy 2007;82:330–9.
[20] Le Pen C, Priol G, Lilliu H. What criteria for pharmaceuticals
reimbursement? Eur J Health Econ 2003;4:30–6.
[21] Meyer F. Outlook and Challenges for the Future: A French Perspective.
Paris, France: Haute Autorité de Santé, 2011.
[22] Zhang H, Zaric GS, Huang T. Optimal design of a pharmaceutical pricevolume agreement under asymmetric information about expected
market size. Prod Oper Manag 2011;20:334–46.
[23] Rizzo JA, Zeckhauser R. Generic Script Share and the Price of BrandName Drugs: The Role of Consumer Choice. Cambridge, Massachusetts,
NBER Working Paper 11431.
[24] Magazzini L, Pammolli F, Riccaboni M. Dynamic competition in
pharmaceuticals: patent expiry, generic penetration and industry
structure. Eur J Health Econ 2004;5:1598–618.
[25] Lexchin J. The effect of generic competition on the price of brand-name
drugs. Health Policy 2004;68:47–54.
[26] Schmid M, Wickler F, Maloney KO, et al. Boosted beta regression. PLoS
One 8:e61623.
[27] Dylst P, Simoens S. Generic medicine pricing policies in Europe: current
status and impact. Pharmaceuticals 2010;3:471–81.
[28] Vogler S. The impact of pharmaceutical pricing and reimbursement
policies on generics uptake: implementation of policy options on
generics in 29 European countries–an overview. GaBI J 2012;1:
93–100.
73
[29] Danzon PM, Chao LW. Cross national price differences for
pharmaceuticals: how large and why? J Health Econ 2008;19:
159–95.
[30] Dalen DM, Strom S, Haabeth T. Price regulation and the generic
competition in the pharmaceutical market. Eur J Health Econ
2006;7:208–14.
[31] Corstjen M, Carpenter M. From managing pills to managing brands.
Harvard Business Rev 2000 March-April.
[32] Grabowski HG, Vernon JM. Brand loyalty, entry, and price competition
in pharmaceuticals after the 1984 Drug Act. J Law Econ 1992;35:
331–50.
[33] Rizzo JA. Advertising and competition in the ethical pharmaceutical
industry: the case of antihypertensive drugs. J Law Econ 1999;42:
89–116.
[34] Berndt ER, Danzon PM, Kruse GB. Dynamic competition in
pharmaceuticals: cross-national evidence from new drug diffusion.
Manag Decis Econ 2007;28:231–50.
[35] Spurling GK, Mansfield PR, Montgomery BD, et al. Information from
pharmaceutical companies and the quality, quantity and cost of
physicians’ prescribing: a systematic review. PLoS Med 2010;7:
e1000352.
[36] Siddiqui M, Vincent Rajkumar S. The high cost of cancer drugs and
what we can do about it. Mayo Clin Proc 2012;87:935–43.
[37] Gagnon MA, Lexchin J. The cost of pushing pills: a new estimate of
pharmaceutical promotion expenditures in the United States. PLoS
Med 2008;5:e1.
[38] Espin J, Rovira J, Olry de Labry A. WHO: working paper 1: external
reference pricing. WHO/HAI Project on Medicine Prices and Availability.
2011. Available from: http://www.haiweb.org/medicineprices/05062011/
ERP%20final%20May2011.pdf. [ Accessed September 25, 2012].
[39] Petrou P, Talias M. A pilot study to assess feasibility of value based
pricing in Cyprus through pharmacoeconomic modelling and
assessment of its operational framework: sorafenib for second line
renal cell cancer. Cost Eff Resour Alloc 2014;12:2.
[40] Danzon PM, Wang YR, Wang L. The impact of price regulation on the
launch delay of new drugs – evidence from twenty-five major markets
in the 1990s. Health Econ 2005;14:269–292.
[41] Claxton K, Buxton M, Culyer T, et al. Value based pricing for NHS drugs:
an opportunity not to be missed? BMJ 2008;336:251–4.
[42] Greiner W. Germany’s drug pricing after AMNOG – What comes next?
Presented at: 7th American & German Healthcare Forum. Minneapolis,
June 21, 2011.
[43] European Surveillance of Antimicrobial Consumption. ESAC yearbook
2006. Available from: http://app.esac.ua.ac.be/public. [Accessed
September 8, 2013].
[44] National Institute for Health Care Management Research and
Educational Foundation (NIHCM). Changing Patterns of Pharmaceutical
Innovation. 2002, Washington, USA.
[45] Espín J, Rovira J, García L. Experiences and impact of European risksharing schemes focusing on oncology medicines. EMINET 2011.
[46] Russo P, Mennini FS, Siviero PD, Rasi G. Time to market and patient
access to new oncology products in Italy: a multistep pathway from
European context to regional health care providers. Ann Oncol
2010;21:2081–7.
[47] Toumi M, Jarosławski S. Market access agreements in Europe: typology
and rationale. Presented at: ISPOR 13th Annual European Congress.
Prague, Czech Republic, November 6–9, 2010.
[48] Goeree R, Jing He, Daria O’Reilly, et al. Transferability of health
technology assessments and economic evaluations: a systematic
review of approaches for assessment and application. Clinicoecon
Outcomes Res 2011;3:89–104.