Drug Pricing in a Regulated Market

�

Dipartimento di Scienze Economiche

Facoltà di Economia Via S. Faustino 74/b

25122 BRESCIA

e-mail: [email protected]

Drug Pricing in a Regulated Market

Stefano Capri� and Rosella Levaggi�

Address for Correspondence: Prof. Rosella Levaggi Dipartimento di Scienze Economiche Università di Brescia Via S. Faustino 74/b 25122 BRESCIA [ITALY]

� ISIS Research, Milano. � Dipartimento di Scienze Economiche, Università di Brescia.

2

Drug Pricing in a Regulated Market

Abstract In this article we analyse the problem of determining the price for new drugs in a market where a stringent budget constraint on public expenditure exists and we suggest an innovative methodology to set their prices. The market is characterised by asymmetry of information and a high proportion of investment in R&D, an element that has to be taken into account because it is only through research that it is possible to obtain better drugs. Our proposed method allows to set the price of new drugs in different market contexts, i.e. where less effective alternatives are already sold or in new markets. We also propose a unified methodology to evaluate the social value of drugs in different markets through the definition of a function of the cost per QALY that society is willing to pay. Keywords: Drug value, price setting, public health care expenditure

3

1. Introduction

When a market can work according to the rules set out for perfect competition there is no need for

regulation because a first best allocation of resources, where consumers maximise their utility and

firms their profits, is reached automatically. Prices on these markets reflect the marginal benefit

from consumption and the equilibrium is found at a point where the marginal cost of production is

equal to marginal utility of consumption. This condition also implies zero profit for the firms and

cost minimisation. The markets for drugs is not a perfect competitive structure for several reasons:

• on the consumption side, the demand does not reflect the marginal benefit of consumers. The

demand for drugs is usually made by a doctor who acts as an agent for his patients hence the

demand might reflect also his preferences; furthermore if the patient has an insurance coverage,

the price he pays is lower than the market price;

• on the production side, drugs have a high proportion of cost in R&D hence the price is usually

much higher than their cost of production.

In reviewing the different regulation policies applied to the drug market we can find at least five

categories of restrictions:

• patent restrictions on commercialisation of the new drug which allows the company to sell new

drugs in a regime of monopoly;

• restrictions on their price ranging from the Italian case where the price and its increase must be

authorised by a government agency to the less restrictive Australian rule to reimburse the

consumer with the price of the cheapest alternative;

• restrictions on the drugs that can be sold, i.e. before a new drug can be put on the market, its use

must be approved on the basis of its effectiveness in controlled clinical trials1;

• Restrictions on the reimbursement scheme. In most countries once a drug is approved to be sold

on the market, its cost is not necessarily met by the State. In some cases the reimbursement

scheme depends on the special circumstances of the patients while in other systems drugs are

classified in different categories according to their importance2;

• budget limits on the purchase or prescription of drugs such as the indicative prescribing amount

set up in Britain for GP's or the double budget procedures in Italy3.

�

1 The rules to be followed to allow new drugs to be sold are not uniform. The most important schemes are described in Capri et al. (2001). 2 In Italy, for example, there are different groups: A whose cost is fully covered by public expenditure, B which consists of prescribed medicines for which the patient must make a 50% copayment, and C that includes drugs for which the consumer has to make a full payment. For a detailed presentation see Levaggi and Capri (1999) cap.7

4

In this article we would like to present the problems for a benevolent regulator who has to set the price

for new drugs in a market where a stringent budget constraint exists. The market is characterised by

asymmetry of information and a high proportion of investment in R&D, an element that need to be

taken into account because it is only through research that it is possible to obtain more effective drugs.

In the field of public economics the literature has recently proposed new models to determine a

socially optimal price in the presence of monopoly and asymmetry of information4. Laffont and Tirole

(1993), for example, set up a framework to examine regulation in several environments from the point

of view of a benevolent regulator who wants to maximise the welfare of his community. In the case of

regulatory policy, the three elements that need to be taken into account are consumer’s surplus, the

profit of the firm and the cost of public funds needed to enforce the regulation5.

These models proposed are however not suitable for the pharmaceutical industry because in this

market both costs and consumers' surplus are difficult to be measured.

In this article we propose an innovative method based on the evaluation of the social value of each

drug. Our analysis stems from the Gravelle (1998) model, but is innovative in many aspects. The

article will be organised as follows: in section two we present the models suggested by the literature to

deal with regulation in the drug market, in section three we present our novel approach and finally in

section four we draw the conclusions of our analysis.

2. The models suggested by the literature

The market for drugs is a typical example of a non-competitive market where a first best allocation

of resources cannot be reached. In this market, in fact, the producer is not a price-taker and can set a

price higher than the marginal cost of the product he is selling6 and government intervention is

needed to correct for this market failure. Several models have been proposed to correct this market

failure, but most schemes fail to reach their objectives because they are based on the assumption

that the costs of the regulated industry can be observed7.

Loeb and Magat (1979) suggest a regulation mechanism to overcome the problem related to the

asymmetry of information on the cost side of the pharmaceutical industry. They propose to use a

system based on the evaluation of the consumer’s surplus: the firm producing an innovative drug is

given a subsidy that is equal to the total consumer surplus generated by the new product. In this case,

3 See Levaggi (1998) for a description of the system. 4. See Laffont and Tirole (1993) and Armstrong et al.(1994). 5 The latter element should be weighted by their shadow price, i.e. the opportunity cost of using funds in this environment. 6 See Tirole (1988) for a complete exposition. 7 See Tirole (1988). Some economists have proposed innovative models that take account of asymmetry of information

5

the firm will sell its product at a price equal to its marginal cost, but the total profit of the firm is

increased by the subsidy. The mechanism automatically secures that the firm pursues a cost

minimisation strategy. This approach is not easily implemented because the demand for new drugs is

difficult to measure, and this information is necessary to evaluate the amount of the subsidy. Gravelle

(1998) summarises the shortfalls of this approach as follows:

• pharmaceuticals needs relevant investments in R&D that is usually a sunk cost for the industry;

• on the research side competition is quite high and it is often necessary to grant patents before the

exploitation of the new drug ;

• the promotion of a new drug plays an important role in its market success; an optimal incentive

scheme ought to take account of this aspect as well;

• the new products might be a substitute for other less effective drugs already available. The actual

improvement needs to be certified in a rigorous way;

• the demand for a drug does not reflect patient’s benefit or willingness to pay since it derives from

an agency relationship between the patient and his doctor.

Gravelle suggests to use a different method, EPV (Ex Post Value of reimbursement), which is based

on the social value of the new drug and is not measured using demand as in a standard microeconomic

approach. The incentive scheme foresees the payment of a lump sum to the firm only if the new drug

is actually sold on the market and a matching subsidy paid on each unit of product sold. The

distinction between the lump sum and the matching grant element is important to give the firm an

optimal incentive to invest and promote the new drug; from the point of view of the optimal quantity

to be sold it is only necessary that the total subsidy is equal to the marginal social evaluation of the

new drug. The scheme proposed by Gravelle and its differences with the approach proposed by Loeb

and Magart are presented in figure 1.

< Figure1: The approach by Loeb and Magat and by Gravelle about here >

The horizontal axis measures the quantity of the drug while the vertical axis measures price. The

line DD represents the demand for the drug, which is downward sloping because the market is not

perfectly competitive. Marginal costs are assumed to increase with quantity and are represented by

the curve MC. If the market is not regulated, a firm would produce the quantity that equalises

marginal revenue to marginal cost, Qm, for a price pm which is not socially optimal. Loeb and Magat

on costs; see for example Baron and Myerson (1982)

6

propose to subsidise the firm to make it produce the quantity Ql at the price pl. This is done by

supplying the firm with the lump sum subsidy plDF. Gravelle (1998) notes that in this market the

social benefit of the drug8 might be higher than the willingness to pay of the consumer; in the case

presented in figure 1 the average social benefit of the drug, BM, is greater than the willingness to

pay of consumers (). The optimal social quantity to produce is the one for which marginal social

benefits are equal to marginal cost. This point is represented by (Qg, pg ) hence the subsidy to the

firm ought to be p-pg. The extra profit granted to the firm gives, according to Gravelle, the

incentive to an optimal investment in promotion and R&D.

This mechanism can be successfully applied both in systems where drugs are fully paid by the State

and in contexts where they are financed through copayment from the patient.

However, the implementation of this scheme is difficult because the marginal benefit function for

the use of a new drug cannot be readily estimated. In a public health care system that is dominated

by the public supply the social value of a new drug can be defined using indirect methods as it will

be shown in the following section.

3. A method to determine the maximum allowable price for a new drug

The social benefit of a drug can be readily transformed into its money value through different

procedures, the most important being cost-benefit analysis, but this method can be successfully used

only if the price of a drug has already been defined. The problems for the regulator in the drug

market is very complicated: in the short-run he has to ensure that the price of new drug is similar to

the alternatives already sold on the market and that the benefits/price ratio is as uniform as possible

across drugs; in the long run a method has to be devised to balance the need to pass onto the

consumer the benefits of innovation and the profit to be secured to the pharmaceutical industry in

order to support research in new drugs9.

In our article we will propose an innovative procedure for the following cases:

• price setting for an innovative drug on a market where other alternatives are already sold;

• price setting for an innovative drug on a market where there are no alternatives;

• price realignment for all the drugs on the market.

The first two cases are short-run problem while the last procedure is related to price setting in the

8 The difference might arise from the nature of impure public good of drugs and from the value that society attaches to research 9 See Benzi (1998)

7

long-run.

In our analysis we focus on the methods to be used to set the price in a market where public

regulation imposes a stringent budget constraint on the use of new drugs. In the conclusions we will

show how our methodology can be easily applied to a market where such a stringent constraint does

not exist.

3.1 Price setting for an innovative drug on a market where other drugs are sold

In this market price setting is more simple; in the short run in fact it is possible to use the total cost

of the less effective drug as a proxy for the social evaluation of the therapeutic properties of that

specific treatment10.

In this case, the elements that should be taken into account to determine the price of the new drug

fall into two main categories:

• the effectiveness of the news drug measured in terms of QALY’s or in any other equivalent

measure; its compliance; its side effects and the value that society attaches to the research in new

drugs. The latter element needs to be kept separated from the commercial use of the patent.

Society as a whole has an interest in promoting research since this is the most effective way to

increase the probability of new discoveries;

• the cost savings that can be achieved through the use of the drug. In this category we can include

the decrease in direct medical cost associated with the use of the new treatment which, for

example, might not need to be used with other drugs and also the savings deriving from less days

of hospitalisation, the decrease in indirect medical costs such as social community care. It could

also be included in this category indirect economic cost saving deriving, for example, from

losing less working days.

To present our methodology, we start with a very simple model; we consider a market where one

drug, F1, producing only one characteristic - approximated by the number of QALY11 gained - S at

rate β112

, is sold for a price pF1. We define a therapeutic unit as QF1 which is equal to the quantity of

F1 that is needed for the drug to be effective. In the short run, a fixed budget constraint R is set for

the purchase of F1. This budget constraint implicitly defines the number of patients that can be

treated. The price of the innovative drug can be determined as follows:

10 This system is implicitly used in countries such as Australia where the less costly alternative is reimbursed. In some countries such as Italy the principle is that a drug with the same effectiveness, same posology and method of administration should be sold at the same price. 11 QALY stands for Quality Adjusted Life Years. This measure is discussed at length in Williams (1992) 12 The production in terms of restored health has to be linked to the units of therapy that can be bought, as it will be

8

1. definition of quantity that can be purchased given the specific budget constraint: 1

*1

FpR

F = .

2. Definition of the number of patients, PAT, that can be treated given that QF1 is the number of

doses that are necessary for a treatment: 1

*1

FQF

PAT = 13.

Once the number of patients has been identified, we can define the total amount of S that can be

obtained with drug F1 as follows: PATS 1* β= and the price of a single treatment

as:111 * FFQ QPP = .

The parameter β reflects the effectiveness of the drug to restore patient's health and it is affected by

different factors such as the probability of success of the therapy and its effectiveness.

In this market a new drug, F2, is introduced which produces S at rate β2>β1. The increased

productivity for β2 can be determined by a greater effectiveness of the new drug, by a higher

probability of success or by both factors.

In the short run we suggest a very simple procedure to set the price for F2; we argue that its

maximum price should be set to the level that leaves society indifferent to buy the old or the new

drug. Society is indifferent between the two alternatives if their supply allows to get to the same

level S* of characteristics.

Given the assumption of a higher productivity of the new drug, we can obtain the same level of S

treating a smaller number of patients (Number Needed to Treat NNT):

2

*

2 βS

NNTF =

The price for each treatment will then be equal to PQF2 = R/NNTF2 while the price of the each unit of

F2 depends on the number of units that are necessary for each treatment, i.e. 22

*2 * FF NNTQF = and the

price should be equal to:

*2

2 FR

PF =

This method can be made more realistic by including the costs associated with the use of the drug. In

our analysis we will use only the direct medical costs cF1 and are assumed to be proportional to the use

of the drug and to the number of patients treated. With this new assumption, the total budget constraint

shown in this paragraph. 13 The effectiveness of a drug depends from the specific characteristics of the patients and by its the recommended use. With QF1 we define the recommended use, i.e. the course of therapy that is necessary for the drug to be effective. The actual effectiveness depends on the specific characteristics of the patients and this is an element that can be observed

9

for the use of drug F1 is:

PATcRR FT 1+ =

where RT is the total cost for the use of drug F1.

Let us now consider the introduction on this market of F2 whose indirect costs are equal to cF2 and

whose effectiveness is equal to β2 .The procedure to evaluate the cost is similar to the one presented

above; first of all it is necessary to consider the number of patients that ought to be treated with the

new drug to get the same level of S*:

2

*

2 βS

NNTF =

Once the number of patients has been determined, we can find the maximum price of each treatment

as follows:

222

FT

Q cNNT

RP

F−=

and from this equation we obtain the price of the drug with the information on the number of units that

are necessary for each treatment:

*2

22 F

PP FQ

F =

If the cost for the use of the new drug is lower than for those existing on the market, it is possible to

shift a part of the budget used for the supplementary cost to finance the purchase of the drug. This

might be a shortcoming of the model since the application of this procedure depends on the sources

of finance; on the budget setting procedures ; on the type of costs considered.

If the direct costs considered in cF1 are the purchase of other drugs, the procedure is correct since the

increased budget for the purchase of F2 is compensated by the reduction of the expenses for other

drugs. If the cost considered derives partly from other drugs and partly from hospital or community

care, the problem is more complicated especially in a public health care system. If the costs for

supplying health care are partly borne by the State and partly by the consumers, it might be correct

to weight the savings with the opportunity cost of the funds; this element does not substantially

change the results presented above, it just makes the solution more complicated from an applied

point of view since it is necessary to evaluate this shadow cost and use it to weight the cost savings.

The example presented in table 1 might be useful to illustrate the methodology. In the short run a

budget of 1000 units is set to buy F1 that produces S at a rate equal to β1= 3. Given a price of 4, 250

only in statistical, average terms.

10

units can be bought; each treatment needs 10 units hence the number of patients treated is 25 and the

total number of QALY produced is 75 . For each patient treated, there is an extra cost of 20 so that the

total budget RT is equal to 1500. In the table we consider two cases: the first one in which the total

budget constraint for the drug includes supplementary costs and a second case in which these costs are

not considered in setting the price to compare the difference in the results.

The new drug that is introduced on the market produces S at a higher rate and presents a lower

direct cost. With the new drug it is possible to reach 75 QALYs treating less patients as shown by

the lower value of NNT2 and the price for F2 can then be higher. If the costs associated with each

treatment are not considered, there is a saving from the total budget RT equal to 143.75. This saving

is due to the joint effect of the 10% reduction in the cost of each treatment and from the reduction in

the number of patients that is necessary to treat to get the same level of QALY gained.

<Table1: Price setting for a drug in a market where already exists a competitor about here >

The problem considered here is rather different from the approach used by the traditional literature on

innovative drug 14 where the price is set and the consumer ought to choose which drug to buy. In our

case we have to define the price of the new drug and for this reason the solution proposed differs form

the standard literature.

The pricing procedure presented above can be easily extended to the case in which many

characteristics determine the social value of a drug. In this case, the maximum price of the new drug is

determined with reference to the level of utility enjoyed by society with the old drug . The procedure to

follow is illustrated for the case in which the social welfare function is linear in the characteristics of

the drug and for a market where a cost cF1 is associated with the treatment of each patient. This

assumption allows to obtain a neat algebraic formulation, but it does not alter the results of the

analysis. The social welfare function is then defined over the characteristics of the drug and can be

written as:

21 )1( =SW SSMax αα −+

where α is the relative weight attached to each characteristic of the drug and S1 and S2 are two

characteristics of the drug which are important to determine its social value.

With a fixed public expenditure budget constraint R and a price equal to pF1 it is possible to buy *

1F doses of drug which, given the productivity levels β1 and β3 and the number of doses per each

14 See, for example, Zweifel and Breyer (1997)

11

treatment allows to treat a specific number of patients (PAT) so that society can reach the following

index of utility:

PATPAT

SS

31*

21*

)1( = SW

)1( = SW

βααβαα

−+

−+

In this market a new drug F2 is introduced. Its production in terms of characteristics is equal to β2 and

β4 respectively while the price for each treatment is equal to cF2. We can now determine the number of

patients to treat, NNTF2 which are necessary to get the same result. From the previous equation we can

write that:

42

*

)1(2 βααβ −+= SW

NNTF

The price of the new treatment will then be equal to:

22

2FQ c

NNTRT

P −=

22

2FQ c

NNTR

P −=

and from this equation we obtain the price of the drug as:

2

2

2

F

QF D

PP =

An example is now presented to clarify the methodology. Table 2 presents the introduction of a new

drug with two characteristics in a market where a drug was already sold. In the example we consider

two alternative cases, the first one in which there is a cost of 20 for each course of treatment an the

alternative case in which there is a 10% reduction in the cost associated with the use of the drug. The

evaluation of the price of the drug is done using α equal to 0,4 which means that the first characteristic

has a relatively lower weight.

As we can note, F2 is more productive in terms of both characteristics and for this reason it can be

granted a higher price. However, the new drug requires a longer treatment; if we do not consider

associated costs its unit price will slightly less than for the drug already on the market. When

indirect costs are internalised F2 has a higher price since all the cost savings can be taken into

account. It is then interesting and important to remark that the consideration of the cost to be

included in the analysis is quite important for the results to be obtained.

<Table2: Price setting in a market where more than one characteristic is considered about here >

12

The price setting procedure described has important consequences for the market for drugs. If the

evaluation of the new drug is done on the basis of one characteristic, it is reasonable to assume that

only one of the two drugs, and usually the most effective, will be sold after the introduction of the

innovative product. If the characteristics on which the price is set are more than one, both drugs can

coexists according to the rules set out by the literature on innovative drugs15.

3.2 Price setting for drugs where there are no other alternatives on the market

If a new drug is sold on market where there are no alternative, i.e. this drug is used for the treatment of

an ailment for which there was no known remedy, we can think to set its price in a way to equalise the

average cost per QALY in other sectors of the drug market.

In the short run we can determine the number of patients that can be treated and hence the budget

constraint in terms of the resources that were used up to that moment for the care of that type of

patients, even if a beneficial treatment was not on offer. In most cases, in fact the use of palliative care

or the "doing nothing" option have costs that, in the short run, can be directed to the purchase of the

innovative drug.

The cost per QALY in each section of the market can be defined as the ratio of the budget to the total

QALY gained; for the first example in the previous section cost per QALY is defined as:

*SR

CQALY =

and a comparison of this cost for drugs already sold allows to determine the average cost per QALY,

QALYC that in the short run could be used as a reference to set the price of the innovative drug.

The technical conditions relating to the new drug allows to define the number of equivalent units of

the new drug that are necessary to gain one QALY. If we assume that the new treatment produces

QALY at a rate β5 and that each treatment requires QF3 doses of drug, the price of each unit of the drug

can be written as16:

3

3

5

F

QALYF Q

CP

β=

while the price for a therapy QF3 is equal to:

15 See, for example, Zweifel and Breyer(14) and Levaggi and Capri (4) cap. 7 16 The following formula can be easily obtained from equations (1) to (3) on the assumption that S* =1, i.e. only one QALY is gained.

13

53βQALYQF CP =

The number of patients can instead be determined on the basis of the budget that is saved from the cost

of unproductive (in terms of QALY) techniques that were previously used to take care of the patients

with that specific disease. If the cost of production for the new drug is higher than the cost per QALY

in other markets, it is possible to grant the innovator the higher price (the so-called premium price),

but this decision ought to be taken using considerations outside the model presented here.

3.3 Price realignment

In the long run the price of the different drugs needs to be revised and realigned. The need for this

revisions stems from several considerations:

• the need to pass onto the patients the benefits of R&D. When the new drug is introduced on the

market, its price ought to be allow the firm to have a good return on its investment in research. In

the long run, however, we can think that part of the benefits from the innovation should be

allocated to the consumer by lowering the cost per QALY gained using the drug;

• drugs in different markets receive different prices per QALY. This might be the result of market

conditions, costs and innovation, but it could also mean that the degree of inertia in the past has

created this inequalities that ought to be reduced.

The methodology we propose can be summarised as follows:

• evaluation of the cost per QALY in terms of the utility enjoyed by society;

• comparison with the actual cost of the drug;

• comparison with the actual cost per QALY.

As concerns the first step we propose to define the cost per QALY that society is willing to pay for

each drug as follows:

)(* QALYgnb

ta

CQALY ++=

where a and b are two parameters, t is the number of years the drug has been on sale, n is the number

of competitors on the market considered and g(QALY) is a function of the number of QALY gained

using that drug. The objective of this formula is to give an incentive to innovation in sectors where

there are few competitors and to introduce new drug. The payment is made up of a lump sum that

depends on the number of years the drug has been on the market and on the number of alternatives that

can be used. The first parameter allows to pass onto the consumer part of the benefits of the

14

innovation. Other things being equal, in fact, the cost per QALY of the drug is inversely related to the

number of years the drug has been sold. The second part of the lump sum is related to the need to

promote research in areas where drugs have few alternatives in order to increase the probability of

diseases to be cured. Finally the last part of the formula is related to the effectiveness of the drug in

terms of number of QALY that can be gained using the formula. The functional form chosen for g(.)

determines this relationship. In general terms we can think that the first derivative of g will be positive

will the second derivative might be positive up to a certain level of QALY to become negative after

this ceiling has been reached. In diagrammatic terms we can express this relationship as follows:

<Figure 2: Cost per QALY gained. About here >

On the horizontal axis we measure the number of QALY gained while on the vertical axis we

measure the allowed cost per QALY. The quantity k measures the lump-sum element, i.e nb

ta +

while the curve kG represents society evaluation of the number of QALY gained for each treatment.

Using this procedure it is possible to define the cost per QALY in different contexts and compare

the gains that can be obtained in different markets for drugs.

Having obtained the theoretical cost per QALY, we can compare this reference value with the actual

cost of production. If this cost is higher than the evaluation of society, the Government can still

decide to introduce the new drug and allow the higher cost. This might be done using considerations

relating to the importance of the innovation or the characteristics of the market in which it is

introduced17.

Finally, if the actual cost per QALY of a drug is different from the evaluation presented above, it is

possible to devise a form of realignment which allows in a set period of time to align its price to the

average on the market. This might be done comparing the actual cost per QALY and the theoretical

one and to set a path to make the two of them equal.

4. Conclusions

In this paper we have proposed a methodology for setting the price for drugs in a public health care

market. The market is characterised by asymmetry of information and a high proportion of investment

in R&D. Our method allows to:

• set a price for an innovative drug on a market where other alternatives are already sold;

15

• set a price for an innovative drug on a market where there are no alternatives;

• align prices for all the drugs on the market.

In the short run, our methodology propose to fix a price that allows to keep consumers’ welfare a

the pre-existing level and to satisfy the constraint of a fixed budget constraint. In the long run, a

realignment procedure can be devised using a pricing formula that takes account of the QALY

gained using the new drug, of the degree of innovation of the product and of its market conditions.

This method might also be used to set prices in a private market. In this latter case the difference

between the short run (price setting for a new drug during the year) and the long-run (price

realignement) is in the evaluation of, n, the number of competitors. For the short run, n should be

evaluated without the new drug, i.e. n should be equal to the number of pre-existing drugs so that

during the year the price for the competitors is not altered by the presence of an innovator and the

innovator receives the same lump-sum as the his competitors.

An important area that the paper leaves unexplored is the actual determination of the parameters of

the cost per QALY function. In this area several alternative are possible and their setting depends on

the objective function of the regulator. The important contribution of our paper is in setting a

common methodology which allows to compare drugs in different markets and that, in the long run,

should be able to reduce the differences in prices that do not depend from specific characteristics of

the drug.

17 This is the case of drugs introduced in markets where there are no alternatives or for the so-called orphan drugs.

16

References Armstrong, M, S. Cowan and J. Vickers (1994) Regulatory Reform, MIT Press, Cambridge, MA Baron, D. P. and R.B. Myerson (1982) Regulating a monopolist with unknown costs, Econometrica, 50(4):911-30 Benzi, G. (1998) New European pharmaceutical system and “drug value”. The European Union Review 3:51-66. Capri S, L. Terranova, F. L, Merlo et al., (and the members of GISF (Italian Group for Pharmacoeconomic Studies) (2001) Guidelines for Economic Evaluations in Italy Recommendations from the Italian Group of Pharmacoeconomic Studies, Drug Information Journal, vol. 35 (in press). Gravelle, H. (1998) Ex post value reimbursement for pharmaceuticals, Medical Decision Making, 18 suppl:S27-S38 Laffont, J.J. and J. Tirole (1993) A Theory of Incentives in Procurement and Regulation, The MIT Press, Cambridge, MA Levaggi, R. (1998) Allocazione di risorse e asimmetria informativa: il caso della spesa farmaceutica, Rivista di Diritto Finanziario e Scienza delle Finanze, 57(1):155-172 Levaggi, R. e S. Capri (1999) Economia Sanitaria, Franco Angeli, Milano Loeb, M. e W.A. Magat (1979) A decentralised method for utility regulation, Journal of Law and Economics, 22:299-404 Tirole J. (1988) The theory of industrial organisation, MIT Press, Cambridge, MA Williams, A. and P. Kind (1992) The present state of play about QUALYs in A. Hopkins (ed) Measures of the quality of life, London: Royal College of Physicians, 21-34 Zweifel, P. e F. Breyer (1997) Health Economics, Oxford University Press, Oxford.

17

Table1: Price setting for a drug in a market where already exists a competitor.

Market before the introduction of the new product

Costs not considered Costs considered

R 1000 1000 S* 75 75 β1 3 3 PF1 4 4

*1F 250 250

Q1 10 10 PAT 25 25 PQ1 40 40 CF1 20 20 RT 1500 1500 Drug value for F2 β2 4 4 NNTF2 18.75 18,75 QF2 10 10

*2F 187,5 187,5

pF2 5,333 6,1 PQF2 53,33 61 R 1000 1143,75 CF2 19 19 RT 1356,25 1500

18

Table2: Price setting in a market where more than one characteristic is considered

Market before introduction of New product

Costs not considered Costs considered

R 1000 1000 S1 75 75 S2 125 125 α 0,4 0,4 SW* 105 105 β1 3 3 β3 5 5 PF1 4 4

*1F 250 250

QF1 10 10 PAT 25 25 PQF1 40 40 CF1 20 20 RT 1500 1500 Drug value for F2 S1 42 42 S2 147 147 β2 2 2 β4 7 7 CC 105 105 NNTF2 21 21 QF2 10 10

*2F 210 210

PF2 4.761 5.342 PQF2 47.61 53.42 CF2 19 19 R 1000 1122 RT 1378 1500

19

Figure1: The approach by Loeb and Magart and by Gravelle

20

Figure 2: Cost per QALY gained.

�

Dipartimento di Scienze Economiche

Facoltà di Economia Via S. Faustino 74/b

25122 BRESCIA

e-mail: [email protected]

Brescia, 27th June 2001 Editor-in-Chief, Prof. Thomas W. Teal Drug Information Journal 501 Office Center Drive, Suite 450 Ft. Washington, PA 19034-3211 USA Dear Prof. Teal I am sending you herewith enclosed three copies of the article Pricing policies the pharmaceutical sector that I would like to be considered for publication in the Drug Information Journal I am looking forward to hear from you. Best wishes, Prof. Rosella Levaggi