Essay 2: Orphan drugs.

This assignment is based on research by dr Kamphorst and dr Karamychev at

the ESE. It consists of an essay in which you will solve and interpret a model. In
addition there are three questions in the main text. Answers to those questions
are not part of the essay, but should be included in the same document, placed
above the actual essay. So they are not part of your essay and do not count
against the word limit.1

1 Orphan Drug Prices

Consider a rare disease. So rare that it is not certain that pharmaceutical
companies are willing to develop a cure, because they expect too little demand.
Consequently, governments may step in to stimulate the development of such
drugs, especially if the disease has debilitating e¤ects or is life threatening. Such
diseases are sometimes called orphan diseases, and any drugs developed to cure
them are orphan drugs.

Question 1: what are two commonly used ways in which the gov-
erment stimulates investments in medical research?

There is widespread concern among the public and policymakers about drug
prices. According to America’s Health Insurance Plans, almost half of the stud-
ied 150 specialty medications cost more than 100,000$ per patient per year.
This exceeds the budget of most households, and a considerable group of pa-
tients cannot a¤ord to follow prescribed treatments (Madden et al, 2008). Prices
are especially high for medication treating rare and serious diseases, such as or-
phan drugs. Messori et al. (2010) and Medic et al. (2017) show that prices
for such drugs are inversely related to the prevalence of the disease. This is
somewhat strange. Why would the rarity of the disease a¤ect its price?
1 There is much to learn about the economics of the pharmaceutical industry. If you are

particularly interested in this topic, Lakdwalla (2018) provides an excellent overview.

1
 Figure 1. Yearly cost per patient vs Number
of patients for 17 non-cancer orphan drugs in
Italy. Source Messori et al. (2010).

One explanation, often put forward by pharmaceutical companies, is that

these drugs are expensive because they need to recover their previous invest-
ments in a smaller market (Bosanquet et al., 2003). So if you half the demand,
you need to double the price to recoup the R&D investments.

Question 2: what is the main problem of this explanation? (If you

don’t know, check the …rst chapter and …rst lecture again)

A related, better explanation is that …rms are less willing to develop a drug
for a smaller market, as demand is lower. This results in fewer …rms partic-
ipating in the market, less competition and, thus, higher prices. We are not
fully satis…ed with this explanation for two reasons. First, for orphan drugs,
most markets are a monopoly (Simoens, 2011), so there is little variation in the
amount of competition. Second, if the market entry decision indeed causes the
inverse relationship, you would also expect a positive correlation between drug
prices and development costs. However, Kesselheim (2016) does not …nd such a
positive correlation.

We believe that there is also a third explanation. In our research, we show

that the inverse relation may be due to optional health insurance which covers
that drug. What does this mean? Consider Table 1.2 Table 1 shows the lifetime
prevalence rates for various diseases. In it, you see that probability of someone
getting Multiple Sclerosis (MS) is less than 1/500 (= 0.002). Suppose that a
monopolist o¤ers a cure for Multiple Sclerosis (MS). Our research shows that,
if MS is a serious enough disease, the price for this cure will be more than 500
times as high when the drug is covered by insurance than when it is not! For
breast cancer, a much more common disease, the price would ’only’be 8 times
as high.
2 This table was constructed by Bram Horstink, who was a RA (research assistant) to dr

Kamphorst.

2
 Disease LTP Source
Breast Cancer 0.125 Feuer et al. (1993)
Congenital Heart Disease 0.01311 (children), Marelli et al. (2014)
0.0612 (adults)
Active epilepsy 0.004 MacDonald et al. (2000)
Parkinson’s Disease 0.002 MacDonald et al. (2000)
Amyotrophic Lateral Sclerosis 0.0008 –0.0043 Johnston et al. (2006)
Multiple Sclerosis 0.00189 Eaton et al. (2010)
Crohn’s Disease 0.00230 Eaton et al. (2010)
Pemphigus 0.00007 Eaton et al. (2010)
Primary Biliary Cirrhosis 0.00013 Eaton et al. (2010)

Table 1. Lifetime Prevalence (LTP) for selected diseases.

Below, we o¤er you a simple model based on our theory. It is able to explain
an inverse relationship between the prevalence of a disease, and the price of the
drug curing it. Your task is to solve the model, and use that as an explanation
of why prices for orphan drugs may go through the roof.3

2 Model
Consider a large population of N healthy individuals. Each individual i; i 2
f1; :::; N g ; has a budget wi ; where wi is uniformly distributed on the [0; 1]
interval. Individuals cannot spend more than their budget. Each individual has
a chance 2 0; 21 of becoming ill, i.e. a patient. So is the prevalence of the
disease. Individual i; if healthy, receives utility

uH
i (wi ) = wi : (1)

Question 3: are these individuals risk averse, risk neutral or risk

seeking ?

The utility of an untreated patient i is

uSi (wi ) = uH
i (wi ) s = wi s; (2)
where s is the utility loss due to the disease. We assume that the consequences
of getting this disease are quite severe, namely s > 1. This implies that the
individual’s willingness to pay for preventing this loss exceeds his budget.
There is a cure for the disease, a drug, which is supplied by a monopolist.
The monopolist has zero production cost and sets price p. Buying the drug
cures the disease, preventing utility loss s. The utility of a patient i who buys
the drug is
3 The model in this assignment is kept as simple as possible. That is enough to demonstrate

the theory. Of course, a simple model may be simplistic. Therefore, in our research, we also
developed a very general version of this model, and demonstrated that our conclusions remain
valid.

3
 uD H
i (wi ; p) = ui (wi p) = wi p (3)
The theory we o¤er is about the e¤ect of insurance on drug prices. What
we will do, is compare the case where individuals cannot insure themselves with
the case where they can.
Suppose that insurance, if available, is o¤ered at cost (e.g. the market may
be perfectly competitive, or the insurance may be o¤ered by a non-pro…t …rm),
so that the insurance premium is given by r = p: If an individual is insured
and becomes ill, the insurance company pays for the drug. The utility of an
insured individual i is equal to

uIi (wi ; r) = uH
i (wi r) = wi r:

For simplicity, assume that people who are indi¤erent buy insurance.4
The timing of this model is as follows:

1. Monopolist chooses price p:

2. Nature draws for each individual i his or her budget, wi :
3. If insurance is available, individuals can choose to buy insurance.
4. Nature draws the health of each individual.
5. Uninsured patients can buy the drug.
6. Patients with insurance or the drug are cured.
7. Payo¤s are realized.

2.1 Essay
Write an essay of less than 750 words in which you analyze the model above.
Include a small introduction with what the essay is about and why it is relevant.
Do not describe the model (so assume that the reader knows it). Instead focus
on the analysis. Do the analysis in two steps. Start the analysis by supposing
that individuals cannot insure themselves, and analyse that market. Then allow
for individuals to insure themselves, and see how the outcomes di¤er. You may
use the following questions to guide both parts of your analysis:

Who will buy the drug?

So what is the demand for the drug?
Which price will a pro…t maximizing monopolist set?
Does this price depend on the prevalence of the disease (so on
the size of the market)? If so, how?
4 In this simple model, this makes no di¤erence. In a more general model, it does make a

di¤erence, but not for the main conclusions which we focus on here.

4
 Once you have done the analysis for both cases, you can compare the results.
Explain what these results mean and give the intuition for the results. You
should focus on explaining why these prices explode when insurance becomes
available, and why drug prices depend on when insurance is available.
Finally, discuss. Typically, in a discussion we talk about (i) robustness of
the model, so do the results change if some assumption changes and how; (ii)
scope of the model, so which situations would …t, and which would not; and
(iii) what the welfare and policy implications are: is it a problem and, if so,
what can be done? To do all three is too much for this essay. Focus on just one
point from the following selection.

Patients typically di¤er in how ill they are. Would it matter if we allow
for this? (Hint: Assume that the expected disutility of being ill, E (s) ;
exceeds 1: So E (s) > 1):

Some people are very wealthy. Bill Gates can a¤ord any treatment. What
if a small group of individuals, share , of the population has w > s;
whereas the rest still has its wealth drawn from a uniform distribution
over [0; 1] : (Hint: Suppose …rst that is almost zero. Discuss then what
happens if is large)

In the text, we speak of severe diseases. But individuals can also buy
insurance for diseases which are less bad. Would the model apply to such
diseases too?
Insurance drives up prices, especially in the case of very rare, debilitating
diseases. Is this a point of concern? Try to give some reason why it is,
and why it isn’t.

If you have space left, you can also discuss a second point that was not
mentioned here. This will not be graded, but the TA may give you up to two
bonus points (0.4 in total) if he or she believes that your point is exceptionally
good. The TAs will send us the best points that they encounter. If interesting
and relevant, we may use some of your points in this study, or a follow-up.
Do not forget to answer the three questions from the text. Position the
answers above your essay.

Good luck!

References:
Bosanquet, N., G. Domenighetti, A. Beresniak, JP. Auray, L. Crivelli, L.
Richard, and P. Howard, 2003. Equity, Access and Economic Evaluation in
Rare Diseases. Pharmaceutical Development and Regulation, 1, 151–157, doi:
10.1007/BF03257374.

5
 Eaton, W.W., M.G. Pedersen, H.Ó. Atladóttir, P.E. Gregory, N.R. Rose,
and P.B. Mortensen, 2010. The prevalence of 30 ICD-10 autoimmune diseases
in Denmark. Immunological Research 47, 228–231, doi: 10.1007/s12026-009-
8153-2.
Feuer, E.J., L.-M. Wun, C.C. Boring, W.D. Flanders, M.J. Timmel, T. Tong,
1993. The Lifetime Risk of Developing Breast Cancer. Journal of the National
Cancer Institute, 85(11), 892–897, doi: 10.1093/jnci/85.11.892
Johnston, C.A., B.R. Stanton, M.R. Turner, R. Gray, A.H.-M. Blunt, D.
Butt, M.-A. Ampong, C.E. Shaw, P.N. Leigh, and A.Al-Chalabi, 2006. Amy-
otrophic lateral sclerosis in an urban setting. Journal of Neurology 253, 1642–
1643, doi: 10.1007/s00415-006-0195-y.
Kesselheim, A., J. Avorn, A. Sarpatwari, 2016. The High Cost of Prescrip-
tion Drugs in the United States: Origins and Prospects for Reform. JAMA,
316(8), 858-871, doi: 10.1001/jama.2016.11237
Lakdawalla, D., 2018. Economics of the Pharmaceutical Industry. Journal
of Economic Literature 56(2), 397–449, doi: 10.1257/jel.20161327.
MacDonald, B.K., O.C. Cockerell, J.W.A.S. Sander, and S.D. Shorvon,
2000. The incidence and lifetime prevalence of neurological disorders in a
prospective community-based study in the UK. Brain 123(4), 665–676, doi:
10.1093/brain/123.4.665.
Madden, J.M., Graves, A.J., Zhang, F., Adams, A.S., Briesacher, B.A.,
Ross-Degnan, D., Gurwitz, J.H., Pierre-Jacques, M., Safran, D.G., Adler, G.S.
and Soumerai, S.B., 2008. Cost-related medication nonadherence and spending
on basic needs following implementation of Medicare Part D. JAMA, 299(16),
pp.1922-1928.
Marelli, A.J., R. Ionescu-Ittu, A.S. Mackie, L. Guo, N. Dendukuri, and M.
Kaouache, 2014. Lifetime Prevalence of Congenital Heart Disease in the General
Population From 2000 to 2010. Circulation 130(9), 749–756, doi: 10.1161/cir-
culationaha.113.008396.
Medic, G., D. Korchagina, K.E. Young, M. Toumi, M.J. Postma, M. Wille,
and M. Hemels, 2017. Do payers value rarity? An analysis of the relationship
between disease rarity and orphan drug prices in Europe. Journal of Market Ac-
cess & Health Policy 5(1), 1299665, 1-16, doi: 10.1080/20016689.2017.1299665.
Messori, A., A. Cicchetti, and L. Patregani, 2010. Relating price determi-
nation to disease prevalence. BMJ, 341:c4615, doi: 10.1136/bmj.c4615.
Simoens, S., 2011. Public health and prevention in Europe: is it cost-
e¤ective? Journal of Pharmaceutical Health Services Research, 2(3), 151–155,
doi: 10.1111/j.1759-8893.2011.00052.x.