Probability Informaton and Risk Communicaton
Lecture 1 – What is Risk Literacy?
Different kind of risks
Dread risks low probability events, but when they happen the cost is really great. Such as,
‘Killer bees kill texas couple’.
Everyday risks risks that are communicated everyday, such as’ if you do this, you have a
higher chance of dying’ etc.
“increased the risk of colorectal cancer by 18%”
Where does the fgure of 18% come from, and what does it mean?
A relatve increase in lifetie risk of developing colorectal cancer.
Approximately 5% of people develop it over their lifetme.
What could a 20% increase in risk mean?
It could mean many different thinks. This is why you should be very sceptcal of assuming risks. They
are very ambiguous.
“Relative risk is usst another way of lying.””
Relatve risks are
The most common way of expressing an increase or decrease in risk.
Prone to exaggeratng an effect; relatve risks are always greater than absolute risks.
Ambiguous because they obscure how great the risk is in absolute terms.
A sign that you should queston the informaton and look further.
Risk literacy is the ability to understand and assess the risks that we face. Risk literacy includes the
ability to interpret informaton about everyday risks, and the ability to communicate these risks to
others in a transparent way.
Risk literacy
Requires statstcal thinking.” This probably differs to what you imagine what you hear the
term “statstcall.
Rests on basic, straightorward concepts which make things simple and clear rather than
coiplex and fog.g.y”
The ostcoies of a diag.nostc test: error
,When you to a test, for example for a disease, these are the possible outcomes
True positve you have the disease, and the test is positve
True negatve you don’t have the disease, and the test is negatve
False positve you don’t have the disease, but the test is positve
False negatve you have the disease, but the test is negatve
Sensitivity and specifcity
The sensitivity is the proporton of people with the disease that are correctly identfed.
To design a test with 100% sensitvity you should say that everyone has the disease.
The specificity is the proporton of people without the disease that are correctly identfed.
To design a test with 100% specifcity you should say that no one has the disease.
There is an approximately a 4% (1 in 26) chance that I don’t have HIV. 4% of low-risk people who test
positve will not be positve.
, The Need for Risk Literacy
We face an overwhelming fow of informaton about everyday risks.
Example Given a positve HIV test, is it certain that you’re positve?
Illusion of certainty nothing is certain, including HIV tests.
Diagnostc tests always trade-off sensitvity and specifcity.
The positve predictve value of the test is what we need to know.
For a low risk individual, there is approximately a 1 is 26 chance of a false positve.
Lecture 1 – What is Risk Literacy?
Different kind of risks
Dread risks low probability events, but when they happen the cost is really great. Such as,
‘Killer bees kill texas couple’.
Everyday risks risks that are communicated everyday, such as’ if you do this, you have a
higher chance of dying’ etc.
“increased the risk of colorectal cancer by 18%”
Where does the fgure of 18% come from, and what does it mean?
A relatve increase in lifetie risk of developing colorectal cancer.
Approximately 5% of people develop it over their lifetme.
What could a 20% increase in risk mean?
It could mean many different thinks. This is why you should be very sceptcal of assuming risks. They
are very ambiguous.
“Relative risk is usst another way of lying.””
Relatve risks are
The most common way of expressing an increase or decrease in risk.
Prone to exaggeratng an effect; relatve risks are always greater than absolute risks.
Ambiguous because they obscure how great the risk is in absolute terms.
A sign that you should queston the informaton and look further.
Risk literacy is the ability to understand and assess the risks that we face. Risk literacy includes the
ability to interpret informaton about everyday risks, and the ability to communicate these risks to
others in a transparent way.
Risk literacy
Requires statstcal thinking.” This probably differs to what you imagine what you hear the
term “statstcall.
Rests on basic, straightorward concepts which make things simple and clear rather than
coiplex and fog.g.y”
The ostcoies of a diag.nostc test: error
,When you to a test, for example for a disease, these are the possible outcomes
True positve you have the disease, and the test is positve
True negatve you don’t have the disease, and the test is negatve
False positve you don’t have the disease, but the test is positve
False negatve you have the disease, but the test is negatve
Sensitivity and specifcity
The sensitivity is the proporton of people with the disease that are correctly identfed.
To design a test with 100% sensitvity you should say that everyone has the disease.
The specificity is the proporton of people without the disease that are correctly identfed.
To design a test with 100% specifcity you should say that no one has the disease.
There is an approximately a 4% (1 in 26) chance that I don’t have HIV. 4% of low-risk people who test
positve will not be positve.
, The Need for Risk Literacy
We face an overwhelming fow of informaton about everyday risks.
Example Given a positve HIV test, is it certain that you’re positve?
Illusion of certainty nothing is certain, including HIV tests.
Diagnostc tests always trade-off sensitvity and specifcity.
The positve predictve value of the test is what we need to know.
For a low risk individual, there is approximately a 1 is 26 chance of a false positve.
