Reflection Paper on these two chapters

A 900–word summary of the two chapters in the module. Be aware that your work will be scanned for plagiarism. Chapter 12 and 14

CHAPTER 12
PHARMACEUTICALS AND THE ECONOMICS OF INNOVATION

Bhattacharya, Hyde and Tu – Health Economics

Intro

The pharmaceutical industry got its start in 1899, when Bayer, a German chemical company, introduced a painkiller called aspirin

Today, the pharmaceutical industry is massive but tightly regulated

This industry is an ideal setting to study both the economics of innovation and the economics of regulation.

Bhattacharya, Hyde and Tu – Health Economics
Ch 12 | Pharmaceuticals and the economics of innovation
THE LIFE CYCLE OF A DRUG

Bhattacharya, Hyde and Tu – Health Economics

3

The life cycle of a drug
Find chemical compound that might treat a disease
Then, test it on animals to show it is not toxic
Then, test on humans in three phases
Phase 1: low dose to healthy individuals (~2 years)
Phase 2: dose to unhealthy individuals (~2 years)
Phase 3: test effectiveness in preventing disease or medical conditions(~3-4 years)
Get approved for sale by FDA or similar body

Bhattacharya, Hyde and Tu – Health Economics
The life cycle of a drug
Once the drug is approved for sale, the drug company has a temporary legal monopoly protected by a patent (17 years in the US)
This is the company’s chance to recoup the millions of dollars spent on testing
After that time is up, other companies can produce the same drug cheaply and profits decrease sharply

Bhattacharya, Hyde and Tu – Health Economics
Ch 12 | Pharmaceuticals and the economics of innovation
DRUG DEVELOPMENT

Bhattacharya, Hyde and Tu – Health Economics

6

Drug development is costly
Hard to find a promising chemical in the first place
Only 21.5% of drugs that enter Phase I pass to Phase III
The whole process can cost $500 million or more to bring a drug to the point of approval

Bhattacharya, Hyde and Tu – Health Economics
Ch 12 | Pharmaceuticals and the economics of innovation
PATENTS

Bhattacharya, Hyde and Tu – Health Economics

8

How do we induce companies to make these costly investments?
Patents create a legal monopoly and hence the opportunity for monopoly profits
In practice, only the top 30% of drugs pay for themselves

Bhattacharya, Hyde and Tu – Health Economics
How strong should patents be?
Downside of stronger patents
Customers have to pay monopoly prices for a longer period
Less incentive for further innovation by same company
Legal barriers to subsequent innovation by another company
But if patents are too weak, no incentive to develop new drugs!

Bhattacharya, Hyde and Tu – Health Economics
Patents in developing countries
Low-income countries think about this tradeoff differently
Monopoly prices weigh more heavily on low-income populations
Free rider effect: if the US has patent protections, companies will develop new drugs even if there are weak patent protections in India
We will see this effect again when we talk about price controls
Price discrimination
In theory, drug companies could sell their drugs for different prices in different countries
In practice, black-market importation makes this impossible

Bhattacharya, Hyde and Tu – Health Economics

11

Price controls
Price ceilings set or negotiated by the government
Example: Italian government publishes list of maximum permissible prices for each drug
Example: NHS in UK sets the price at which they are willing to purchase drugs (monopsony power)
Controls reduce incentive for research
Tradeoff between access to existing drugs and incentives to develop new ones, same as patent tradeoff
US has no broad price controls
Other countries free-ride on US market for future innovation
Drug companies count on making their money off US consumers

Bhattacharya, Hyde and Tu – Health Economics
Ch 12 | Pharmaceuticals and the economics of innovation
INDUCED INNOVATION

Bhattacharya, Hyde and Tu – Health Economics

13

Induced innovation
Definition– discoveries that result when innovators change their research agenda in response to profit opportunities
Example: changing demographics
As the US population aged between 1970-2000, drug companies turned their attention to drugs for the elderly (glaucoma medication, etc)

Bhattacharya, Hyde and Tu – Health Economics
Induced innovation
Academic and public institutions also engage in induced innovation, even though they do not usually profit directly from their discoveries.
Examples:
Large-scale production of penicillin during WWII by US Department of Agriculture
In recent years, academic researchers have focused more on obesity
US Army ceasing malaria research after Vietnam War ended and troops came home from malarial regions

Bhattacharya, Hyde and Tu – Health Economics
Who is harmed by induced innovation?
Diseases that are rare (orphan diseases) or that mostly occur in developing countries (tropical diseases) receive less attention from researchers, because there is less profit to be made.
Governments have tried to harness the power of induced innovation to fight these diseases
Orphan Drug Act in the US
Advanced purchase of yet-undiscovered vaccines for HIV, malaria, TB

Bhattacharya, Hyde and Tu – Health Economics
Ch 12 | Pharmaceuticals and the economics of innovation
REGULATION OF THE PHARMACEUTICAL INDUSTRY

Bhattacharya, Hyde and Tu – Health Economics

17

FDA regulation ensures safety of drugs
Before 1930s, no safety or testing regulations for drugs in the US “magic elixirs”
Thalidomide & Europe (1960)
Prescribed to pregnant women with morning sickness
Caused birth defects in over 10,000 newborns
Pulled from shelves and promoted stricter drug regulation throughout the world
Kefauver-Harris Amendment in the US (1962)
Companies must prove new drugs are safe and efficacious through clinical trials
Stricter regulations led to a lower number of new chemical entities on the market (Peltzman 1973)

Bhattacharya, Hyde and Tu – Health Economics
Type I and Type II errors
The FDA has to decide how restrictive to be when approving new drugs
Restrictive vs. Permissive regulations
Type I error = bad drug is approved (e.g. Vioxx)
Type II error = good drug is rejected or delayed (e.g. beta blockers)
Phase III trials do not have complete information about a drug

Bhattacharya, Hyde and Tu – Health Economics
There is a tradeoff between rejecting good drugs and approving bad drugs
Choosing T* will always lead to some error
ROC plots the tradeoff between Type I and II errors
Regulators balance social welfare and potential harm
More incentive to avoid type I errors because of media attention
Type II errors rarely get in the media because they are hard to catch

Bhattacharya, Hyde and Tu – Health Economics
Other regulations
Doctors have prescription power
True in most countries
benefit: less intentional and unintentional abuse of drugs
cost: time, inconvenience, expense
Bans on direct-to-consumer (DTC) advertising
Bans in place in most developed countries except US
benefit: prevent moral hazard, reduce strain on doctor-patient relationships
cost: customers may not find out about new drugs that will benefit them

Bhattacharya, Hyde and Tu – Health Economics
Conclusion
Tradeoffs
Patents incentive for innovation vs. affordable prescriptions
Government price controls innovation vs. affordability
FDA Regulation more new drugs vs. fewer dangerous drugs
Type I and II errors approve bad drugs vs. decline good drugs
Doctor prescriptions increase safety of drug use vs. expensive drugs

Bhattacharya, Hyde and Tu – Health Economics

CHAPTER 14
HEALTH TECHNOLOGY ASSESSMENT

Bhattacharya, Hyde and Tu – Health Economics

Intro

Health technology assessment (HTA) is comprised of two parts:

Cost effectiveness analysis (the science of comparing the costs and benefits of different medical treatments)

Cost-benefit analysis (the process of choosing an optimal treatment by creating a tradeoff between money and health)

HTA may sound dry and technical but it generates enormous controversy because it involves placing an explicit value on human life.

Bhattacharya, Hyde and Tu – Health Economics
Ch 14 | Health technology assessment
COST EFFECTIVENESS ANALYSIS

Bhattacharya, Hyde and Tu – Health Economics

3

Cost effectiveness analysis
Definition: the process of measuring the costs and health benefits of various medical treatments, procedures, and therapies.
Cost effectiveness analysis (CEA) is the less
controversial part of HTA, because it is concerned
with measuring costs and benefits, not balancing
them against each other.

Bhattacharya, Hyde and Tu – Health Economics
Cost effectiveness analysis
Often multiple treatments, with varying costs, can be used to treat a given disease.
In such cases
How do insurance companies decide which treatments, if any, to provide coverage for?
How do patients decide between an expensive and highly effective treatment and a low-cost treatment that is less effective?

Bhattacharya, Hyde and Tu – Health Economics
Cost effectiveness analysis
If one treatment is both cheaper and more effective than a second treatment, then the second treatment is said to be dominated by the first.
It is never optimal to use a dominated treatment, because there is always a more effective and cheaper alternative available.

Bhattacharya, Hyde and Tu – Health Economics
Cost effectiveness analysis
If neither treatment is dominant, one treatment must be both more expensive and more effective.
In such cases, cost-effectiveness analysis is used to help people decide whether the extra expenditure is worth it.

Bhattacharya, Hyde and Tu – Health Economics
Incremental cost-effectiveness ratio (ICER)
Consider two treatments for the same disease: A and B. A is both more expensive and more effective than B, so neither treatment dominates the other.
The ICER of using A over B is:

Bhattacharya, Hyde and Tu – Health Economics
Lead poisoning example
Which treatment strategy is superior?

Bhattacharya, Hyde and Tu – Health Economics
Lead poisoning example

This ICER provides a price for avoiding a reading disability.
In some sense, people can avoid a reading disability for an average price of $7,241.
Note that the ICER does not make a determination about whether this is worth it or not, it is just an empirical fact about costs.

Bhattacharya, Hyde and Tu – Health Economics
The average cost-effectiveness ratio (ACER)
Q: So why not just look at the various treatments’ ACERs and pick the one with the lowest cost per additional year of life?
A: ACERs typically will not reveal all the potentially cost-effective drugs.

Bhattacharya, Hyde and Tu – Health Economics
Ch 14 | Health technology assessment
THE COST EFFECTIVENESS FRONTIER

Bhattacharya, Hyde and Tu – Health Economics

12

Cost-effectiveness frontier (CEF)
Definition: a subset of treatment strategies for a condition that are not dominated by any other treatment. Any treatment on the CEF is said to be potentially cost-effective.
The CEF simplifies comparisons between treatments by allowing analysts to rule out dominated drugs (which should never be used), and focus only on options that potentially cost-effective.

Bhattacharya, Hyde and Tu – Health Economics
EX: Consider possible treatment options for the disease “bhtitis”: A, B, C, …, I
Cost-effectiveness frontier (CEF)

Bhattacharya, Hyde and Tu – Health Economics
Cost-effectiveness frontier (CEF)
Connect non-dominated options to form CEF

Bhattacharya, Hyde and Tu – Health Economics
The slope of the CEF between two points is equal to the inverse of the ICER between the two.
Cost-effectiveness frontier (CEF)

Bhattacharya, Hyde and Tu – Health Economics
Ch 14 | Health technology assessment
MEASURING COSTS

Bhattacharya, Hyde and Tu – Health Economics

17

Measuring costs
In order to calculate an ICER, we need to measure the costs of each treatment.
Whether a treatment is found to be cost- effective depends upon the perspective taken, because treatment costs and benefits differ for each party.
The social planner perspective: all costs count.
The patient perspective: only costs directly borne by patients count.

Bhattacharya, Hyde and Tu – Health Economics
Which costs count?
Suppose a complete course of a new lung cancer treatment costs $1,000. Is this the only cost to consider?
What if…
the treatment must be administered in a distant location, or for extended periods of time?
the treatment is uncomfortable—or has unwanted side effects?
the treatment will lead to adverse health effects in the future?

Bhattacharya, Hyde and Tu – Health Economics
Which costs count?
How should future costs be counted? If lung-cancer patients are cured but then go on to have costly heart attacks, should those costs count against the treatment?
There is active debate about which kinds of future costs should be included.

Bhattacharya, Hyde and Tu – Health Economics
Ch 14 | Health technology assessment
MEASURING EFFECTIVENESS

Bhattacharya, Hyde and Tu – Health Economics

21

How is “effectiveness” measured?
One common measure of effectiveness is increased life expectancy.
But how do we account for other health benefits that affect quality of life (e.g. increased mobility and freedom from pain)?
The Quality-Adjusted Life Years (QALY) approach combines quality of life and life expectancy into a single index.

Bhattacharya, Hyde and Tu – Health Economics
QALYs
In a QALY calculation, each year of life receives a quality weight q between 0 and 1 that reflects the quality of that life-year.
A year lived in perfect health has a quality of weight of q = 1.
Maybe a year with chronic cough and insomnia is only worth q = 0.5, or a year confined to a wheelchair is only worth q = 0.25.
Who has the right to make this judgment? We will return to this question

Bhattacharya, Hyde and Tu – Health Economics
QALEs
Calculating QALYs requires estimating three pieces of information:
the probability Pt of surviving to each year t
the quality of life qt for each year
a time-discount rate (usually between 3% and 5%)
A person’s quality-adjusted life expectancy (QALE) is the number of additional years he expects to live, weighted by the discounted quality of his life in each of those years (i.e. the sum of his QALYs).

Bhattacharya, Hyde and Tu – Health Economics
Survey methods: quality weights
Visual analogue scale (VAS) asks respondents to rate health outcomes between 0 (worst) and 100 (best)

Pros: simple to administer and easy for respondents to understand
Cons: does not require respondents to think about tradeoffs between different health states. Thus, results may not reflect the intensity of respondents’ preferences.

Bhattacharya, Hyde and Tu – Health Economics
Survey methods: quality weights
Standard Gamble (SG): For health condition H, respondents choose between having H with certainty or a gamble with probability p of full health and probability (1- p) of death. The point of indifference between these two options is used as the quality weight q of health condition H.

Pros: reflects intensity of preferences better than VAS.
Cons: this approach may be affected by risk aversion, and people often respond in counterintuitive ways to such uncertain gambles.

Bhattacharya, Hyde and Tu – Health Economics
Pros: Reflects intensity of preferences better than VAS.
Cons: may be biased if Ŧ* is a function of age
Survey methods: quality weights
Time trade-off (TTO): respondents choose between 1) living for t years with a health state H before dying, and 2) living for a shorter amount of time Ŧ in full health before dying. The quality weight q of health state H is the ratio Ŧ*/t (with Ŧ* representing the point of indifference between the two options).

Bhattacharya, Hyde and Tu – Health Economics
Whose opinion matters in QALY surveys?
Healthy survey respondents may be unequipped to imagine the quality of life in health states they have not experienced.
Expert panels are unlikely to ably represent patients’ preferences.
People who have lived with a condition for decades tend to understate the suffering that healthy people would feel if they suddenly developed a condition (e.g. blindness).

Bhattacharya, Hyde and Tu – Health Economics
Ch 14 | Health technology assessment
COST-BENEFIT ANALYSIS: PICKING THE OPTIMAL TREATMENT

Bhattacharya, Hyde and Tu – Health Economics

29

Cost-benefit analysis (CBA)
Definition: Cost-benefit analysis (CBA) is the process of choosing an optimal treatment among all potentially cost-effective ones, given a certain monetary value for each unit of health effect.
This optimal treatment is then termed cost-effective for a person or agency with that valuation.

Bhattacharya, Hyde and Tu – Health Economics

When we place a monetary value on each QALY, we implicitly create a set of indifference curves that can be plotted with the CEF (why?)
Example: Let us assume that a person values each QALY at $100,000. As a result, his indifference curves slope such that he is indifferent between one additional QALY and $100,000.
We will revisit the crucial question of how to value life-years accurately.

Cost-benefit analysis (CBA)

Bhattacharya, Hyde and Tu – Health Economics

Plot the indifference curves
Find the tangency point
With these indifference curves, the cost-effective treatment is Drug C.
Cost-benefit analysis (CBA)

Bhattacharya, Hyde and Tu – Health Economics
The impact of insurance
Consider the same situation, but now with an insurance package that covers 90% of drug costs.
This insurance coverage lowers a patient’s out- of-pocket price per QALY. With insurance:
Drug C now costs 10% of $160,000 = $16,000 for 3 QALYs
Drug H now costs 10% of $360, 000 = $36,000 for 3.4 QALYs.

Bhattacharya, Hyde and Tu – Health Economics
Moral hazard and HTA
With insurance, the person chooses Drug H, whereas without insurance, he would have opted for the cheaper Drug C.

Bhattacharya, Hyde and Tu – Health Economics
Moral hazard and HTA
The additional costs of Drug H are shared by every one in the patient’s insurance pool.
Thus, insurance encourages technology overuse and the development of inefficient innovations.
Given the problem of moral hazard, many insurance companies use rationing, in order to limit such technology overuse.

Bhattacharya, Hyde and Tu – Health Economics
Rationing
Definition: any method for allocating a scarce resource other than prices.
Example: Insurance companies or national health systems may decline to pay for certain treatments that are not cost-effective.
In the case of our CEF, the insurance company might decline to cover the expensive Drug H.
As the insurance company is only willing to cover 90% of expenses for Drugs A and C, patients would likely respond by choosing Drug C.

Bhattacharya, Hyde and Tu – Health Economics
Rationing
While moving from α to β leads to worse health and fewer QALYs, the money saved on health care costs can now be spent elsewhere.

Bhattacharya, Hyde and Tu – Health Economics
Ch 14 | Health technology assessment
VALUING LIFE

Bhattacharya, Hyde and Tu – Health Economics

38

Estimating the value of life
The CEF helps us find the best treatments for a given budget, but it does not indicate how much we should spend to obtain more life years or QALYs.
In other words, how much is a QALY worth?
To answer these questions, value of life estimates rely primarily on three sources:
labor market choices, product purchase decisions, and government policies.

Bhattacharya, Hyde and Tu – Health Economics
Estimating the value of life
You may think life cannot be valued economically or has an infinite value. Consider the following example:
There is a suitcase across a busy street with a million dollars in it.
If you cross the busy street to get the suitcase, there is a 1% chance you will be struck by a bus and killed.
Do you risk it?
If you answer yes, your life cannot be worth more than $100 million to you ($1 million divided by 0.01).

Bhattacharya, Hyde and Tu – Health Economics
Using the labor market to reveal VSL
In order to attract workers to more hazardous jobs, high-risk employers offer additional wages (“risk premiums”), which supplement the wages workers would earn in comparable, but lower-risk jobs.
If researchers know both the risk premium for a job and the difference in risks, then they can estimate how much a worker values his life.
Example: A worker who would take a job with a 1% higher fatal injury risk for $50,000 more in wages has a VSL of $50,000 ÷ 0.01 = $5 million.

Bhattacharya, Hyde and Tu – Health Economics
Using purchase decisions to reveal VSL
Example: Jenkins et al. (2001) used price data for children’s bike helmets to estimate their VSLs.
The decision to wear a helmet indicates a judgment that the risk reduction of head trauma from bike accidents is worth the cost of buying helmets.
Researchers used the prices of helmets to estimate a lower bound for the value of risk reduction and use that to calculate a lower bound for the VSL of helmet-wearers.
Other purchasing decisions: smoke detectors, safe cars vs unsafe cars.

Bhattacharya, Hyde and Tu – Health Economics
Using government policies to reveal VSL
Example: In 1972, a U.S. law guaranteed kidney dialysis to all patients under 65 with end-stage renal disease for free.
Kidney dialysis costs approximately $50,000 per QALY.
The passage of this amendment suggests that a QALY is worth at least $50,000 to American taxpayers.

Bhattacharya, Hyde and Tu – Health Economics
Conclusion
Insurers can neither cover every single new technology, nor refuse to cover all new procedures
Instead, most insurers are selective about which procedures to cover
HTA is a tool that many insurers and national health systems use to make these coverage decisions
In the policy lectures ahead, we will hear a lot more about HTA

Bhattacharya, Hyde and Tu – Health Economics

Chapter 14: Cost-Effectiveness Analysis | 387

In the case of health insurance, though, customers do not make socially optimal decisions be-
cause prices are distorted. Moral hazard justifies cost-effectiveness analysis, which in turn neces-
sitates research into quality weights and life valuation. Without such research, companies and
policymakers could never figure out which treatments are worth covering from a social planner’s
perspective.

Although rationing is understandably contentious, it might nevertheless be optimal for insur-
ance customers. Rationing can drastically reduce overall health costs and allows for lower pre-
miums (or a lower tax burden in the case of public insurance). Recall Figure 14.8, which depicts
the tradeoff between moral hazard and risk. The contract at ↵ is full insurance and induces the
most moral hazard; this is the case where an insurance company is willing to cover any treatment,
including ones like Drug H that do not produce much health benefit. The contract at � is partial in-
surance with rationing, perhaps a contract that does not cover cost-ineffective treatments like Drug
H but still covers treatments like Drug C.

Figure 14.8: Rationing and moral hazard.

fullness

health care
expenditures

full

↵ (no rationing)

� (rationing)

The indifference curves in this figure suggest that customers are actually happy to live with
rationing in insurance. Rationing reduces the fullness of their insurance because there are some
moments – when they catch bhtitis, for instance – when rationing will prevent them from enjoying
extra years of life they could have had with Drug H. But rationing also saves customers from
having to pay for everyone else’s ineffective drugs. Sometimes, those savings are enough to make
rationing appealing. Moving from ↵ to � leads to worse health and fewer QALYs, but the money
saved on health care costs can be spent on other things that are more valuable.

In order to actually decide whether this decrease in health expenditures is worth it, we need
some measure of the value of life. Figure 14.6 depicted indifference curves that implied that people
valued each quality-adjusted life year at $100,000. If this figure underestimates the true value of
life, socially-optimal treatments will not be covered. If this figure is an overestimate, then there will
be substantial harm from moral hazard.

14.6 Valuing Life
How much is your life worth to you right now? Perhaps you think it is infinite or cannot be valued
economically. No amount of money could convince you give up your life right now. Similarly, you
would never take even the slightest risk like interacting with a sick person or driving a car. If your
valuation of life is truly infinite, you would make some odd decisions. For example, suppose an