Module 3: Aging and Spending
Week 7: Sources of information, where to find quantitative information and why it matters
• Health care systems differ and change in quantitative ways (cost per capita, cost in % of GDP;
share of hospital, physicians, prescription drugs in total costs; number of doctors, nurses, and
hospital beds per 1000 people; number of admissions per year and per person; average length of
hospital stays; and number of visits per year to a physician)
• The health care system differs not only between different countries, but also changes drastically
• There are 2 centralized databases where you can find all the information pertaining to health care
systems most helpful to Canada: the OECD and CIHI
• Based on the fact that health care costs increase and that the elderly cost more, the common
inference is that our aging population is going to bankrupt the healthcare system
• There are 3 main policy recommendations in order to solve this
o Ration care:
▪ Targets high users (high spenders) which indirectly ends up being elderly people (as
they are more likely to be using the most resources)
▪ Some provinces to cut outpatient drug coverage to those over the age of 65
Overall this is an unfair and inefficient way of cutting costs, most of the time
individuals who are using the most health care are using it because they actually need
▪ It is a very good thing that spending is concentrated
▪ In Ontario
• 1/3 of costs are spent on 1% of the population
• 80% of costs are spent on 10% of the population
• Therefore cutting on high spenders means cutting on those who really need
o Privatize care:
▪ Good for public finances but inefficient; with more out of pocket expenses penalize
the sick. The costs don't go away they are just shifted to the sick
▪ The model from this is based on CPP; the Canada Pension Plan where Canadians pay
into the fund while they are working throughout their life
▪ The money from CPP cannot be used for anything besides healthcare
▪ Fair across generations
▪ Can be inefficient: if interest rates are low it won't really benefit
▪ Spending on healthcare is not wasting money
• Before making a decision on what to do we need to actually know if aging will bankrupt the
healthcare system, there are two approaches to that question:
o Naïve and pessimistic: the demographic scenario
o Sophisticated and optimistic: the time-to-death scenario
OECD: An international effort to standardize statistics
• Organization for Economic Cooperation and Development
• For the 35 richest countries in the world
• They work to produce comparable information on economic matters (trade, production,
investment, social policies, migrations, and health) However they do not have information on spending by age, therefore we need to use the next
database more specific to Canada
Session 1, Canada: Open CIHI website and find NHEX information to answer the following questions
1. How much each Canadian is expected to cost in health care in 2016 (on average)?
On average, each Canadian is expected to cost $6,299.
2. How much is Canada as a whole expected to spend on health care in 2016?
As a whole, Canada is expected to cost $228.1 billion.
3. What share of GDP does it represent in 2016?
This will represent 11.1% of Canada's GDP. This is one measure of the total resources we
have. However there are some resources not included in the GDP, it is generally only
4. How does the current trend look like? And what can we expect as a result?
The current trend is down from its peak in 2010, it seems similar to the trend of the mid-
1990s. Since 2010, health spending per capita has decreased by 0.1% per year. 11.8% in
2010, and now it is around 11%
5. What was the share of the GDP in 1975? Comment.
In 1975, the share of the GDP was 7%. Health expenditure was about the grow during this
time period. The number increased by more than 50% from 1975 until now, which is a
6. Which province has the lowest health care cost per capita? And which has the highest? Comment
on the difference.
Quebec has the lowest health care cost per capita at $5,822. Newfoundland and Labrador
has the highest at $7,256. This could be due to different age distributions, population
density and geography. Quebec has a larger population, however Newfoundland's
population is older. Quebec has the second fastest aging population in the world, the largest
in Canada. Quebec saves money on healthcare because they scale their economy by scaling
costs (they don't need as many people to administrative work), and population distribution
(there are more people in larger cities, saving money on health transportation costs).
7. What is the cost per capita of the territory with the highest cost? Why is it so high?
Costs in the territories are so much higher because of their large geographical area and low
population. The highest territory is the Northwest Territories at $15,065. The population is a
lot younger in the NWT, but because the areas people are living in are so remote, it costs
significantly more for health care services.
8. What is the share of seniors (65+) in total health care spending and how does it compare to their
share of the population in 2014?
In 2014, seniors spending on health care averaged at $11,635 per person. This has increased
from 12.6% in 2000 to 15.6% in 2014. 46% of total healthcare spending goes to seniors, and
seniors represent 16% of the population. Which means on average, seniors cost 3 times
more than the average person.
International: Open OECD Health database (from Mac library account) and find the answers to the
1. What was the share of GDP spent on health care in Canada in 1975? Compare this finding to the
NHEX and explain why it is different.
According to the OECD, the share of GDP spent on health care was 6.5% in 1975. It is a big
difference in terms of spending money (basically increasing taxes by 10%). The reason for the different is because the OECD looks at capital expenditure, which excludes capital
spending, whereas the NHEX looks at current expenditure. NOTE:
Current expenditure means that the money spent is immediately gone. Examples: paying
someone, buying food,
Capital expenditure means that they money spent lasts longer, its more of an investment.
Examples include: equipment, buildings, computers/technology.
2. What was the share of GDP spent on long-term care in Canada in 1975 ad in 2015? Comment.
In 1975, Canada spent 0.7% of GDP on long-term care, in 2014 it was 1.4%. Part of the
increase has to due with the increase in longevity.
3. What is the share of GDP spent on health care in Mexico today? Comment on the difference with
In Mexico (2015) the share of GDP spent on health care is 5.8%. In Canada (2015) the share
of GDP spent on health care is 10.1%. This is because Mexico is poorer than Canada, and so
they have to spend more on basic necessities.
4. How many professionally active physicians per 1000 population are there in Canada?
As of 2014, there were approximately 88,873 physicians. The latest update was in 2014,
having 2.61 physicians for every 1000 people.
5. How many were there in 1990? Comment on the difference between 1990 and today.
In 1990, there were 2.11 physicians for every 1000 people. We have approximately 25%
more today than in 1990.
6. How many are there in the US? Comment on the difference between the US and Canada.
As of 2013, the US had 809,845 physicians, which made their ratio 2.70 per 1000 people.
They have more doctors, their healthcare costs significantly more.
1. What are the 3 policy options discussed to prevent population aging from bankrupting the health
Rationing, privatizing (shifting the cost from public to private), and pre-funding (taking
money to put to the side for future funding)
2. Why would rationing target elderly individuals?
Rationing would target elderly individuals indirectly because they tend to be the higher
users and need more visits, hospital stays, or drugs than the rest of the population.
Rationing will set limits on how much an individual patient can use and will target high
3. Discuss the pros and cons of rationing.
Pro: might reduce inefficiencies, services that are used but not necessary or effective in
Con: will penalize those who truly do need care
4. Discuss the pros and cons of privatizing.
Pro: reduces taxes
Con: consequence of rationing, penalizes the sick and poor
5. What is pre-funding of health care? Under what condition would it work?
Pre-funding is a mechanism through which we pay a tax today that does not go to general
revenue but instead into a special fund to be used later once the population is older and
needs are greater. It works well when interest rates are high and the fund can generate
revenues, otherwise it’s a waste of money.
6. Discuss the pros and cons of pre-funding health care. Pro: it is more fair than the current system in the sense that the generation pays today for
the care they will receive tomorrow.
Con: it may prevent governments to invest in current health care (because they don't have
the money). Less investment today can mean slower medical progress
Effect of population aging on health care costs
The Demographic Scenario
• The total cost that we spend in a given year is done by the following calculation: take average cost
of a person of a given age (say age 0 newborns) multiplied by the number of persons in that age
group. Now add the average cost per person for the next age group multiplied by the number of
people in that age category. This trend continues until you add the product of average cost per
person aged 90+ by the number of people in that age group.
• We know that in 2050, there will be more persons in the old age categories (60-64, 65-69, and so
on until 90+). Because the average cost per person in these categories is higher, total cost will also
be higher in 2050 than in 2013
• The DS says that eventually we will shift the spending from younger people to older people,
however in order to do that we need to increase costs on spending.
• DS describes a 'wave effect' (water volume increases where the ground is higher).
• OECD predicted that increase in public health care spending in Canada by 4% of GDP. This would
mean an increase in taxes by approximately 10%.
The Demographic Scenario is wrong
• The age profile of health care spending is not constant because age is not the determinant of cost.
It just so happens that as we age we need more health resources. However if we imagine that
people in the future are going to have better health then we may be able to decrease the costs
The Time to Death Scenario
• The true reason the elderly spend more is they are more likely to be in their last year of life. For
everybody being in their last years of life it means much higher healthcare costs for that time
• If mortality declines, fewer people of a given age will be in their last year of life in 2050 compared
to now. Which will mean that the average cost per person will decrease in each age category, even
more so in older ages. The age profile will be flatter and at a lower level.
• As a result aging through decreased mortality will reduce spending and partially offset the effect
of the baby boomers becoming old.
• Increased longevity will offset the baby boom generation.
• The TTDS works like the DS except that the age profile changes over time. When lower mortality is
in effect then the average cost per person is also lower.
• Using this scenario, the OECD predicts that the increase in public health care spending in Canada
will only increase by 1%, meaning that taxes would only need to increase by 2.5%.
Laura Funk page 33-36
1. What is apocalyptic demography? Apocalyptic demography is about overstating the negative effect of demographic trends for
society. Apocalypse is the end of times, meaning the final catastrophe.
2. Why can it be coined demographic scape-goating?
It can be coined as this term because it can be used to pitch generations against each other
and blame social issues as well as public finance problems on the baby boomers who 'used
up' all the benefits.
3. Why is it used to cut public programs and why does Laura describe it as ideological?
It is used to cut public programs because TDR is lower now than in the 19060s (which was
the peak of the boom). Cutbacks are motivated by market-oriented policies and global
competition and aging is used as an excuse to justify them.
Yang et al. Longevity and Health Care Expenditures
1. What is a survivor and what is a decedent?
A survivor is someone who is not going to die in the next 12 months. A decedent is someone
who is going to die within 12 months.
2. Describe the data used MCBS 1992-1998.
This is longitudinal data collecting health care costs by type of service (hospitals, institutions,
others) at the individual level and by month. 25,000 individuals aged 65 and older are
followed over 5 years max. overall, they have 750,000 person-months (an average of 2.5
years per individual).
3. If an individual dies in March 1996, what is their time until death in 1993?
3 years (36 months)
4. If an individual is observed for the last time in March 1998, at which time they are alive, what is
their time until censoring in March 1993?
3 years, censoring means that they could no longer observe the death of that individual at
the time they stopped the study.
5. Describe the age profile of monthly spending (65-97).
Average cost per capita increases significantly with age.
6. Describe the time to death profile of monthly spending (36-death).
Monthly cost increases slowly from 3 years before death to 12 months ($1500 -> $2000 per
month. It then picks up dramatically around 4 months before death ($3000), and then again
at 1 month before death ($7000). Again, this instills the fact that the in the last year of life,
costs increase dramatically.
7. Comment on figures.
a. 4(A): the age profile of cost among decedents is flat. It costs the same to die, whatever the
age at death.
b. 4(B): the age profile of cost among survivors increases slowly but regularly.
c. 5(C): the cost of inpatient care in the last year of life decreases substantially with age.
Hospitals clearly spend fewer recourses to try and save older decedents than younger ones.
d. 6(C ): and 6(D): both show that the age profile among survivors is the result of nursing home
care, which the cost does not increase with proximity to death but rather increases with
Health Care, Longevity, and Prolongevity
Effect of Spending on Aging
The Tsunami Effect
• Health care costs increase o Ex: in 1975 Canadians spent $527 on healthcare per person; in 1995 that increased to
$2534; in 2014 that increased again to $6069.
o Even after controlling for inflation the numbers are still significantly increasing. However we
do know that it is not due to population aging, but rather the Tsunami effect: We spend
more over time for individuals at the same age.
This is different than the wave effect because the increase is due to per capita spending, not
o The Tsunami effect is driven by income.
▪ At a given age, we are just going to spend more. That is the reality, when we become
richer, we spend more.
▪ Beyond that, we spend a larger portion on our GDP on healthcare. This is due to:
• Changes in utilization/volume (seeing doctors more often, spending more time
in the hospital, etc.)
• Changes in unit costs: price of a physician visit doubled in 40 years, and price in
a hospital stay multiplied by 9.
▪ In order to understand this shift we need to understand the technical progress in
Technology and the Expansion Effect
Technical progress in health
• More progress in cure than care
o Care is the task of observing and providing support. The only technical innovation here is
ITC or robots and evidence-based decision making tools.
o Cure is the task of actually performing something, it can be more easily supplemented or
substituted for by machines and robots (cameras) and molecules. Cure is the activity in
healthcare where productivity gains are possible (and potentially large).
• Technical progress does 2 things
o Cut costs per service provided
▪ The substitution effect: instead of paying a surgeon, now have a machine or pill that
prevents even having surgery
o Increases demand for services through lower unit costs and/or better quality.
▪ The expansion effect: it becomes easier to provide the service for more people.
▪ More supply = more demand
• The scale of innovation
o New drugs, new molecules, new devices
o What we have now did not even exist 50 years ago
o There has been a diffusion of new technology and treatments
o Expansion of indication: a drug that already exists can be used for multiple things
o Intervention: the organization of care, the way doctors do things, etc.
o Innovation is defined as high technology, not necessarily expansive
▪ Understanding the biological mechanisms of disease is highly technological but
treatments are not
▪ Ex: Vaccinations
▪ It takes a lot of knowledge and research in order to be able to create these high tech
solutions to health problems.
• We understand that we won't get more treatments out of better technology but rather overcome
diseases, age healthier, etc. Determinants of Longevity
• First Period 1750-1950
o Income strong determinant of longevity at the individual/social level
o Increase in life expectancy will widen social inequalities in health: those who can access
clean water (in low income countries) and medical treatments (in high income countries)
will increase their longevity but not those who cannot.
o However income is not the only determinant, i.e. women are generally poorer than men but
o There is also not a unique driver over time, known as the Preston Curve
Shows a relationship at the country level between GDP per capita and life expectancy
▪ More importantly, it shows that the relationship is not stable over time
o Gains in life expectancy were made at young ages, but now they are shifting to being made
at older ages.
o True causes of the first period
▪ Nutrition: having better nutrition would likely lead to less infectious diseases
Sanitation: the main driver, includes things like draining swamps, filtering and
chlorinating water, pasteurizing milk. This has eliminated countless deaths from
things like typhoid, cholera, dysentery, etc.
• Second Period (recent period) 1950-now
o Since 1960, cardiovascular disease mortality has declined by over 50% (which accounts for
70% of the 7 year increase in life expectancy between 1960 and 2000)
o 2/3 of this decline can be attributed to medical treatments, therefore 45% can be attributed
to medical progress
o Overall, we can buy longer and healthier lives by spending more on medical treatments.
Discussion of Aubrey de Grey
• Why do we need to defeat aging?
We need to defeat aging because it kills people.
• Why are we resisting the idea?
We are resisting the idea of aging because its not just about life its about healthy life,
getting old isn't fun but rather a global trance. People resist the idea of aging because it
would be boring, they wouldn't be able to afford it, and dictators would rule forever, and it
would be impossible to maintain a normal fertility and mortality rate, because as mortality
increases fertility has to decrease. There are risks associated with radically extending lives,
however, by not doing the research we are impacting future generations by not giving them
the choice of living longer.
3. Why is it unethical to resist it?
It is unethical to resist aging because of the issue of children. People say that if we defeat
aging then what will happen to people who want to have children? People would have to
decide whether they wanted to have children or live longer. It is also unethical for us to be
"imposing our values on the future" just because today many people think that extending
the lives of humans is wrong, doesn't mean future generations won't, and therefore we
should still be putting the research in now.
4. What is the difference between biogerontologists and geriatricians? Biogerontologists focus on metabolism, and working on prevention. Geriatricians get
involved once pathology begins, so once aging begins they try to stop it.
5. What would be the role of the engineers and why would that work better?
Engineers would work better because they specifically work on the metabolism that leads
to pathology. They would repair the parts of the body that needed to be repaired in order
to maintain the threshold.
6. What is the difference between breakthroughs and incremental progress?
Fundamental breakthroughs have a longer time frame in terms of predictability.
Incremental progress normally happens as an after effect of a fundamental process. After
breakthroughs, it’s a catalyst for expanded improvements. Once you get through the
breakthrough the process is a lot more rapid.
7. Is we find a breakthrough that adds 30 years of healthy life at age 55, now, how many years of
extra life will the 55 years old get?
They wont just live 30 yrs longer, it's a much larger increase because as they are
experiencing that 30