simple tools for understanding risks (from innumeracy to insight)
DESCRIPTION
TRANSCRIPT
![Page 1: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/1.jpg)
Simple tools for understanding risks: from innumeracy to insight
An article by:
Gerd Girenzer and Adrian Edwards
Presented by:Muhammad Husnul Khuluq
F131877
![Page 2: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/2.jpg)
Haemoccult Test
for Colorectal
Cancer
The prevalence
of cancer was
0.3%, the
sensitivity of the
test was 50%,
and the false
positive rate was
3%
What does it mean?
DO I have a cancer?
Bad Presentatio
n
Confused/ Mispercept
ion
Inappropriate Action
Confusing numerical
representations
1. Single event probabilities
2. Conditional probabilities
3. Relative risks
![Page 3: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/3.jpg)
Single event probabilitiesThis medicine,
however, might have 30% to 50%
chance of developing a
sexual problem
Oh, NO! I’ll have 30-50% sexual
problems.
Different reference classes (in mind) results in ambiguity
Of every 10 people who take fluoxetine, three to five will experience a sexual problem
Better Representation
Specify the reference class
Use frequency statementOR
![Page 4: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/4.jpg)
Conditional Probabilities
The probability that a woman has a breast cancer 0.8%.
If she has breast cancer, the probability that a mammogram will show a positive result is 90%.
If a woman does not have breast cancer, the probability of a positive result is 7%.
If a woman has a positive result, what is the probability that she actually has breast cancer?
8 out of 1000 women have breast cancer.
Of these 8 women with breast cancer, 7 will have a positive result on mammography.
Of the 992 women who do not have breast cancer, some 70 will have a positive mammogram.
If a group of women have positive mammograms, how many of them actually have breast cancer?
sen
sit
ivit
ysp
ecifi
cit
yp
osit
ive
pre
dic
tiv
e
Conditional probabilities Natural frequencies
Presenting three conditional probabilities is likely becoming a trouble for understanding the
information
Natural frequencies refer only to the same class of observation. Therefore, confusion could be diminished.
![Page 5: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/5.jpg)
Conditional probabilities VS natural frequencies
The doctors who were given natural frequencies tend to estimate more precisely than those who were given conditional probabilities
![Page 6: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/6.jpg)
Relative Risks
Women aged over 50 years are told that undergoing mammography screening reduces their risk of dying from breast cancer by 25%
Women in high risk groups are told that bilateral
prophylactic mastectomy reduces their risk of dying from breast cancer by 80%
Of 1000 who DO NOT undergo mammography 4 will die from breast cancer
Of 1000 who DO mammography, 3 will die
from breast cancer
Of 100 who DO NOT undergo mastectomy, 5 will die from
breast cancer Of 100 who DO mastectomy, 1 will die from breast cancer
Absolute risks reduction
People tend to misunderstand the relative risks like (for the 1st case), of 1000 who do the test, 25% will save and 75% others will die from breast cancer
X
![Page 7: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/7.jpg)
Confusing Factors and How to Avoid it
1. Reference classFor each case of confusing numerical representation, misunderstanding of the reference class always exists
Clarify the reference class first.OR
Use transparent representations like natural frequencies
2. FramingNegative framing (negative-oriented information) tend to worry the patients very much and negate to undertake the treatment
Use negative frame for communicating clinical risk that the patient might still choose whether to take or not
Use positive frame to persuade patient for risky treatment
3. ManipulationThe use of verbal explanation tend to confuse the patients
Manipulate the way of representing statistical information.
The use of pictorial representation or chart might be helpful for patients to understand the information
![Page 8: Simple tools for understanding risks (From innumeracy to insight)](https://reader036.vdocument.in/reader036/viewer/2022082323/5457af0db1af9f33608b4f00/html5/thumbnails/8.jpg)
Strength Weakness
The article is well structured, completed with examples, and charts
The important point are highlighted, so people will not loose the content
Less explanation of the example given. So it becomes less
readable for people who are not familiar with the terms
No explanation why statistical innumeracy is ignored (whereas in the beginning they it is given
as a problem)
Thank you very much for your attention