Week 8 Lecture 1 - Google Docs
Week 8 Lecture 1 - Google Docs
Week 8 Lecture 1 - Google Docs
Midterm
● Tuesday October 29
● 60 questions
● 8:30 - 10:20 am
Agenda
● Context: screening and public policy
○ How we decide what we screen and test for and who pays, etc.
● PKU – the Guthrie Test
● WHO: justification for screening
● Ontario Policy – criteria used for selecting tests
● Research Context: Incidental Findings & Actionability
● Ethical considerations: normality, eugenics, rights, the ”big 4”, human nature (heuristics &
Prospect Theory)
● Examples and applications
Learning Objectives
● To understand the origins of screen test success
● To summarize the ethical issues surrounding screening tests
● To identify the justifications for screening as identified by the WHO
● To identify how screening tests are selected in Ontario
● To differentiate between testing circumstances as part of clinical care and as part of research
● To define “incidental findings” and “actionability” and how they apply in research screening
● To understand how “normality” contributes to successes and drawbacks in screening tests
● To identify examples where screening tests are widely used in clinical practice and how positive
and negative results are interpreted
Genetic Tests have significant legal, social, ethical and public policy implications
● Genetic tests provide information and some level of entertainment (23 and me test kit)
Cool to see where you came from
○
○ But also issues related health and healthcare
4 key issues (important considerations)
● Autonomy
○ The ability to make decisions for yourself as an autonomous and educated human being
who has reached the age of majority and can make choices for themselves
● Confidentiality
○ The right to confidentiality
○ Even if you do get some of these tests done for health reasons or just out of interest, the
default is that your result should be confidential and you are the only person to have the
right to share it unless you have agreed otherwise
○ Confidentiality is NOT the same as privacy
■ Confidentiality = People are given information in the expectation that it will be
held in confidence. It is necessary usually to do their job
■ Ex: if in a health care hospital setting, clinicians have access to your health
record and use that info to understand your background, comorbidities, and help
make the best case decisions
■ Using it and keeping it in a confidential way
○ Different than ANONYMOUS
■ Anonymous is when your results are absolutely unable to be traced back to you
as an individual
● Privacy
○ Privacy is another important consideration
○ Privacy and confidentiality go rather hand-in-hand because privacy is essentially your
right to confidentiality
○ Most of the legislation in Canada refers to privacy of information and protection of it as
opposed to confidentiality
○ Privacy is more the actionable term in our ability to have information about us stored
safely and securely and used appropriately
○
● Equity
○ Genetic tests also have significant social and ethical implications regarding equity
○ A good deal of early genetic work was focused on white people and less research on
visible minorities
○ It is important to offer genetic screening and genetic testing to EVERYONE regardless of
the background or origin status
○ Different ethnic groups experience genetic illness differently -- some groups are unique to
specific illnesses (e.g. nescilamia? or tacsacticy?) where it is a ethnic linked genetic trait
as opposed to something that the general population can have
● These 4 considerations are important on their own, but they're even more important when you put
them together
○ Like toppings on a sandwich
○ T hese considerations are combined with legal contexts of regulatory frameworks for
protection of those things, and social standards and social expectations regarding genetic
testing/screening
● Policy – consensus & critical
○ Then are then of course parts that become public policy in terms of actual government
focused endeavors in this realm
○ The policy realm has a lot of BOTH consensus and critical approaches to genetic
screening of health care
○ Consensus = the nuts and the bolts of how a genetic testing policy happens
■ would include things like:
■ What tests are we covering?
■ Is it males or females that are being tested?
■ Is it all age groups?
■ How will we return the results?
■ What will we do with the data
■ Basically all of the items you would think about that you need to keep your
program running
○ Critical = focused more on the social, ethical implications
■ Include:
■ Are we testing all ethnic groups?
■ Are women and men fairly treated in this realm?
■ What about children?
Ethics
●
○ The ethical considerations
○ Is it ethical to test for something that you can actually do something about?
○ Or is it ethical to identify traits in people with no specific traits in mind?
○ There are these considerations
Key Terms
● Important terms for this content!
● Screening vs Testing
○ Recall: we talked about this with tuberculosis
■ Context varies
■ In TB, about trying to find communicable disease in a population
■ This case is different, looking at the genetic basis for illness or possible illness in
specific populations and for specific purposes
○ Concept is the same for genetic screening and testing
■ The ACTIVITY/concept is the same, but the context is different
● Genetic Screening:performed in the general population; early detection, monitor risk
○ Similar to disease screening
○ It is performed in the general population
○ The intention is to provide early detection of disease or possible disease, and monitor risk
○ You go LOOKING for risk factors = screening
● G enetic Testing: Applies to individuals who are seen as having a higher probability of having a
particular condition; determine presence/absence
○ Once you have SYMPTOMS, you are no longer screening you are TESTING
○ Once determined that someone has a higher likelihood of having a particular condition
(i.e. screening), then you set out to actually make sure beyond a reasonable doubt or to
the best of sciences ability whether someone actually has a condition, yes or no
● Specificity vs. Sensitivity
○ Our reliance on genetic screening/testing, whether it becomes part of healthy public
policy or just public health policy, are the concepts of specificity and sensitivity
○ This is something you can study in biostatistics and epidemiology
○ The distinction between the two is really important!!!
● Specificity: Is the ability to correctly rule out disease
○ The statistical reliability or ability of a genetic test/genetic screening instrument to
correctly
● Sensitivity: Is the ability to correctly pick up positive cases
○ The statistical reliability of genetic tests to correctly identify cases where the person is
positive
● Truth = the status of an individual who has undergone genetic testing/screening
○ They would either find themselves in a binary disease state (either have disease OR don't
have disease)
○ Binary status -- either have a condition or don't have a condition
○ Truth is typically ruled out through testing and by actually finding out what is going on
○ When screening, we are looking for risk factors but we want to make sure that the
screening test is consistent and coherent with the ACTUAL reality or the TRUTH
○ Cases of people with either:
■ With disease
■ Without disease
■ People who definitely don't have a disease and they aren't carriers
● Specific: ability to RULE out disease (i.e. D) - ability of the test or to obtain normal range or
negative results for a person who does not have a disease
○ A person who does NOT have a disease and the test result shows that they don't have the
disease
○ Ex: COVID tests
■ Covid tests in their early days more very specific (and in some case sensitive)
and now that COVID has mutated (and if using expired tests), the ability to be
specific is often not very good
■ Can't rely on an expired test like covid always to be specific because it often has
people with the disease, but identifies them as being negative (i.e. false negative)
○ True negative = D
■ People who are legitimately non-carriers of a disease and the test result
corresponds with that
■ This is SPECIFIC
○ A specific test will always show NON-DISEASE and NON-DISEASE in the TEST
RESULT
● Sensitive: ability to ID positive cases (i.e. A) - proportion of true positives tests out of all patients
with a condition
○ Know how many POSITIVE cases there are
○ They have the condition and the test will ALWAYS show that this is the case
○ Ex: X-Ray or any other radiographic exam
■ A really good example of a sensitive test
■ Obvious break in a bone or tumor can be seen and identified as such
Ex: Breast screening
●
○ Different types of breast tissue are prone to not being specific or not being sensitive
○ The screening is mammography
○ Mammography = done on anyone now over 40 (not limited to females)
○ Not very specific for people with specific types of breast tissue (categorized as A, B, C,
D -- from dense to fatty)
○ If you have dense tissue you need to go back for another test (e.g. ultrasound, CT) that is
more specific and able to rule out disease
● IDEALLY, the best screening test will be both SPECIFIC and SENSITIVE!
○ Some conditions we don't "care" if they are both (bc not required)
○ Ex: if you just want to make sure that people DON'T have COVID, then you just have to
be specific
■ Not necessarily interested in identifying people who do
○ In communicable disease, a sensitive test is important because we want to know who is
sick so we can quarantine them and keep them away
■ The specific part of the test will help identify the inability to have disease
spreading throughout the community
Predictive Value
● In biostatistics and epidemiology, sensitivity and specificity are a lot more common than the
diagram shown previously
○ H ow we gauge the success of any genetic screening/testing is its ability to identify the
positive cases where required and the identification of negative cases where required, and
sometimes both -- done via the PPV or NPV of a test
● Positive Predictive Value (PPV) is a ratio of the true positives (numerator) relative to true/false
positives (denominator)
○ i.e. (total number of A)/(total number of A + B)
○ Positive predictive value = look at the POSITIVE row
○ The higher the number of true positives (compared to false positives), the BETTER the
Positive Predictive Value
○ A ratio of how well a test does at being positive in its prediction
● Negative Predictive Value (NPV)is a ratio of the true negatives (numerator) relative to true/false
negatives (denominator)
○ i.e. (D / [C+D])
○ The opposite of PPV
● Important because if you have a test that is not doing a good job at being reliable in showing true
positive's, there is often no value in using it
○ Or maybe society in the practice of medicine has been using a specific screening test for a
long time and its predictive value is 50%, but with improvements in science and
technology, another screening test will allow you to improve that to 70-75% and then you
might switch tests
○ Sometimes a poor positive predictive value is better than nothing because you don't have
anything, but if you are making a choice as a policy maker, you want to look at what is
the BEST accurate test that we have that has the best positive predictive value, or best
negative predictive value because some conditions are more aligned with that
● Improvement is relative to the costs that is present
○ If science could come up with a really good way to improve mammography screening for
breast cancer, to improve the NPV (specificity) or PPV (sensitivity) and it costs millions
more, the new test would theoretically not be economical/more efficient than the system
that we have
○ Which is why the list of terms at the beginning included "cost-effectiveness analysis"
○ The purpose of predictive value in policy selection is the most economical to do the best
job (i.e. want to do the job without breaking the bank)
○ These are data that a policy maker would use when trying to decide whether they should
adopt a certain genetic screening/genetic test procedure
● Disease prevalencein a population affects PPV and NPV. When a disease is highly prevalent, the
test is better at ‘ruling in' the disease and worse at ‘ruling it out.
○ Disease prevalence is really important in the application of screening and testing whether
it's a disease that is congenital, chronic, or communicable
○ The predictive value is affected by the disease prevalence in the community
○ When a disease highly prevalent disease:
■ Better at ruling IN the disease
■ Worse at ruling OUT the disease
■ Therefore, better off with a test that is more SPECIFIC than SENSITIVE
○ Ex: COVID -- when large numbers of cases are circulating in the community
It is better to use a test that rules it OUT (specific test)
■
■ Specific tests are more important to identify "having disease" vs "ruling it out"
COVID testing program and the rationale behind it
●
○ Scientists have learned the natural history of disease and have a better understanding of
how it behaves, the natural life cycle, etc.
○ Since we KNOW MORE, we don't have to test it like we did 3-4 years ago
○ It is a policy choice
h
● ttps://www.nature.com/scitable/topicpage/human-testing-the-eugenics-movement-and-irbs-724/
● Ex: In WW2 Eugenics became an outlet for a lot of hatred and persecution for many
○ German experiment in WW2
● In the USA, there have been a number of eugenics movement that were attempted in the US
T
● hey have all been looked upon in history as never a good thing
● The quality of the human population for desirable traits -- similar to animal breeders with
agriculture
● Ex: Davenport founding the Eugenics record office laboratory
○ About improving quality
○ Regrettably turned into a movement where a number of members of society were
medically treated so they were unable to have children, certain families with certain traits
were discouraged form having children
○ The whole idea was to remove certain genes from the population
○ Many of these policies included involuntary sterilization or institutionalization
● Identifying normal leads down ethical rabbit holes
○ There are a whole load of ethical considerations and problems associated with race
hygiene
○ Perhaps may have started off as innocent and hopefully helping society but can do the
opposite
When?
● Prenatal
● Newborn
● Among various members of the population (i.e. late onset like Huntington’s)
○ Can screen for older people (40+) -- e.g. breast cancer
○ Huntington's -- can identify that you have it and it will be late onset but treatment avail in
early stages to slow it down
● Cancer Care Ontario
○ 4 screening programs at Cancer Care Ontario
○ Breast screening, lung cancer screening, colon cancer screening, etc.
● Once symptoms appear, no longer screening
● Under-writing: evaluates riskiness of a proposed insurance policy
○ More of an issue in America
○ When trying to get HC, and have an underlying medical condition/problem, they are
excluded from coverage in the policy
○ Underwriting is part of insurance provision in the US
○ It is part of the insurance process in Canada for more private insurance (e.g. life
insurance)
○ Medical questions will identify issues that evaluate your risk relative to proposed
insurance policy
○ Often, life insurance doesn't have a lot of underwriting in younger people
■ Life insurance usually stops once you turn 80 and you can't get it back most of
the time bc the risk of you needing it would be sooner rather than later
● Propensity and availability of treatment
○ Propensity = the high risk target
○ Targeting high risk people and having treatment ready to get people at an early age of
illness is really important part of the policy in Ontario
● As condition of employment? Discrimination
○ Is it important to ask for genetic screening to be done as a condition of employment?
○ Ex: working in coal mine and susceptible to a certain lung condition
○ But can also lead to cases of discrimination in the workplace and unfair hiring practices
due to genetic "baggage"
I n Ontario, there is PRENATAL SCREENING and NEWBORN SCREENING
●
● Prenatal screening
○ Ethical red herring
○ Can find out before baby is born if your baby is likely to have
■ Down Syndrome;
■ Edwards Syndrome
○ Non-invasive -- everyone gets ultrasound
○ Amniocentesis -- but if you are at higher risk, you can have an amniocentesis
■ Test amniotic fluid and test it for more specific and sensitivity
○ Can also find out sex through ultrasound
Ethical Principles
● Related to both ethical review of clinical trial research and normal practice of medicine are the
ethical principles of principlism
○ Important in setting policy AND also implementing policy and making it actionable
every day
○ These principles are key when sharing information that comes about through genetic
screening/testing
1. Respect:
○ individuals should be treated as autonomous agents, and that persons with diminished
autonomy are entitled to protection.
○ In the course of research, people with diminished autonomy that we focus on are
incarcerated individuals, children, and First Nations
○ People who aren't autonomous or autonomy has been restricted in some way are entitled
to protection
2. Beneficence:
○ do no harm, maximize possible benefits, and minimize possible harms.
○ Should refer back to the WHO definitions of when you should to screening
3. Non-Maleficence:
○ prevent agents of harm
4. Justice:
○ fair treatment of all persons; sharing of burdens and benefits
○ Clearly, there is no one “correct” policy..
■ As justice and intersectionality have lots of manifestations in society
5. Autonomy
○ An overarching principle
○ People are individuals and if they have reached the age of majority they should be able to
make their own decisions
Minimizing Harm
● Minimizing harm = non-maleficence and beneficence
○ Screening is never fool proof, which is why we move to testing when we're not sure
○ But sometimes there are confounders and other factors
● Potential downsides may include:
○ Test interpretation:
■ T ests, aimed at healthy people at high risk (e.g. due to family history) are hard to
interpret since many conditions are multifactorial (i.e. so many genes and so
many environmental factors)
■ Healthy people at high risk = propensity is high
■ Society and people within society are complex so it is hard when you are well
and not showing any symptoms to rely on tests
■ Recall: this connects back to the importance of predictive value and specific and
sensitivity in testing
○ Lack of options to treat.
■ There can be a long lag time between linking a gene mutation with a disease and
developing effective à utilitarian arguments and cost-benefit questions.
■ You can screen for something but a long time lag identifying disease and then
developing effective treatment
■ Nothing you can do about it so what's the value in doing that?
Danger to employment and insurance
○
■ Some conditions make people vulnerable to illness getting a disease
■ Certain workplaces can make that worse
○ Psychological impact:
■ anxiety over the results, survivor’s guilt, fears of discrimination and social
stigmatization.
■ Sometimes these impacts may outweigh the benefits of testing
■ Ethical/psychological component of screening
■ The psychological impact is not a consistent value across individuals which is
why we have the right to know and the right to remain ignorant because different
people have different perceptions of risk
Moral Hazard
● Moral Hazard: lack of incentive to guard against risk where one is protected from its
consequences, e.g. by insurance
● In the context insurance and screening and testing, there is an important concept called moral
hazard
1. Moral hazard is the counterargument to "the right to know" and "the right not know"
● Insulating people from risk may make them less concerned with the potential negative
consequences of that risk than they otherwise might be
1. Preventing people from understanding their risks might make them less concerned with
potential negative consequences
2. Ex: know that smoking causes cancer and if we don't tell this to people, they are going to
become less concerned with the potential negative consequences
■ In turn, has implications for HC system and the COST
● Health insurance in US (or Canadian private health insurance) always incorporates moral hazard
into its policy application
● Co-pay or deductible or refusal to provide policies for high risk
1. C o-pay gives you a small portion of the overall cost -- makes you more in tune with the
overall value of what you are getting
2. Deductible
■ Ex: seniors have to pay the first "X" amount of dollars then the rest of their drugs
are covered
3. This puts some of the responsibility of the negative consequences on the individual (e.g.
misuse of emergency departments, engaging in risky health behaviours, etc.)
Often a policy decision to work on moral hazard for either:
●
1. The benefit of reducing overall costs to provide for that aspect of society
2. Or to make people more aware/more careful
Prospect Theory
● Recall: connects to the psychological impact of harm that may result from testing
○ Prospect theory is a way for us to look at that idea
● Prospect theory: people tend to react differently to gains than to losses (risk averse to losses but
less so for gain – e.g. lottery tickets)
○ Ex: They don't want to lose money but happy to buy lottery tickets
● people make decisions on the basis of subjective assessments of probabilities which may be quite
different from the objective or true probabilities.
○ The right to know, the right to remain ignorant, and the critical arguments of moral
hazard all go into prospect theory because people in the HC system make decisions on if
they want to know or not know, based on subjective assessment of probabilities
Which may or may not be different from reality
○
i.e. like framing can alter risk perception (Deber and Mah, 2014)
●
○ Framing, how people have different decision theories and scaffolds, and worldviews are
going to assess risk DIFFERENTLY and usually assess risk based on PROBABILITIES
○ Ex: If statistics say you have "X" percentage likelihood of developing skin cancer if
you've been burnt severely by the sun 2 times
■ People tend to make decisions based on those probabilities
■ Which may or may not be true
● Chart:
○ People make SUBJECTIVE assessments of probabilities
■ Internal assessment
■ People tend to UNDERESTIMATE their likelihood of heart disease, cancer,
and other natural causes by a significant amount
■ People tend to OVERESTIMATE their likelihood of being involved in an
accident, homicide, or other unnatural cause
○ Reality vs Subjective
■ Ex: Someone who smokes a pack of cigs a day may not see themselves as high
risk for lung cancer. But statistical estimate may be HIGHER
● Just based on worldview and framing mechanisms, you are going to see risks differently
○ And most of society does not have access to or care to find statistical estimates that they
face daily
○ Part of their decision theory is related to prospect theory in that they have created
scaffolds or structures as to how they perceive risks and this helps guide them without
them thinking about it, in terms of making decisions relative to the right to know or the
right to remain ignorant
Discrimination and Confidentiality
● Confidentiality means your data is identifiable but it is kept safe and it is ONLY used by the
people who need to use it
● Discrimination can take place if confidential information gets out and is used against you
● People are treated differently by their employer or insurance company because they have a gene
or mutation that increases risk of inherited disorder
○ In the US especially
○ Some of them are external to the place that they work, others not
● USA - GINA; Canada - GNDA
○ To counteract these problems, a policy response to the proliferation of genetic
information and the right to know and not knowing, and more people knowing exist
○ These are two pieces of federal legislation that prevent discrimination on the basis of
genetic information
● USA --Genetic Information Non-Disclosure Act (GINA)
○ GINA prevents insurers from using genetic information to
■ Determine whether someone is eligible for insurance
■ Can't use that info for underwriting (i.e. determining risk which creates the cost
you pay for insurance)
● Canada --Genetic Non-Disclosure Act (GNDA)
○ GNDA in Canada
○ Essentially the same as GINA
○ Protects Canadians from discrimination based on genetic characteristics
○ Genetic info can't be demanded from you, nor can they be used against you
○ GNDA recently survived a constitutional challenge in 2020 (deemed constitutional)
○ In relation to the GNDA, appropriate amendments to the Labour Code andCanadian
Human Rights Act
■ So that all of the policy instruments (in this case legislation or regulatory) are
ALIGNED
■ Important to align legislation across ALL SPHERES so that it's consistent
○ Labour Code amendments
■ Employers cannot demand genetic info, nor can they use it to discriminate
against staff
○ Canadian Human Rights Act
■ Then it becomes written into the Human Rights Act
■ As well as the Ontario Human Rights Legislation to be consistent
○ There is often a delay
■ Gives governments time to impose changes to their legislation to follow the suit
with the federal level
○ GNDA does not apply to family history, diagnosis of manifested disease or symptoms
■ Law doesn't apply to genetic testing related to manifested disease
■ Or family history
● GINA prevents health insurers from using genetic information to determine if someone is eligible
for insurance or to make coverage, underwriting, or premium decisions
● G NDA protects Canadians from discrimination based on genetic characteristics – can’t be
demanded or used against someone
● GNDA survived a constitutional challenge in 2020; amendments to Labour Code and Canadian
Human Rights Act
● GNDA doesn’t apply to family history, diagnosis of manifested disease or symptoms
Screening Newborns
● Congenital Hypothyroidism
○ Pros: Highly sensitive, acceptable specificity, inexpensive, non-invasive small blood
sample
○ Cons: Not preventative (but can administer treatment after testing)
● Sickle Cell Anemia
○ Pros: More prevalent in certain ethnicities, inexpensive treatment
○ Cons: Not all treatment is effective, not prevalent in Canada, less successful prophylaxis
● Krabbe Disease
○ Pros: Life-threatening disease, severe disease complications
○ Cons: Not very prevalent (1/100,000), Expensive treatment and may be without benefit,
potential for over diagnosis
● Duchenne Muscular Dystrophy (DMD)
○ Pros: Selective disease, Highly Specific, Carrier Testing (mothers), highly sensitive
○ Cons: Expensive, Invasive,No treatment, Pregnancy termination issues
● Cystic Fibrosis
Pros: Carrier, non-invasive test, prevents long term health issues, specific
○
○ Cons: Not highly sensitive (false positives common), pregnancy termination
● Deber & Mah, 2014
○ There were only 5 things that they screened for in newborns in 2014 compared to now
where they screen for MANY more
○ In 10 years, the ability of tests to be specific and sensitive, and predictive value has
improved and has changed how screening is done
Policy in this area is not static
●
○ Screening is appropriate when testing continues (this is an example of that)
○ Screening expanded a lot over the last 10 years