Gigerenzer et al
Psychological Science in the Public Interest –Volume 8, Number 2, pp 53 –
- conditional probabilities – which include sensitivity and false positive rate
- better to use natural frequencies
- example
o Conditional probabilities
|
|
|
| 1 woman |
|
|
|
|
| 1% chance of breast cancer |
|
|
| 99% no breast cancer |
|
| 90% positive |
| 10% negative |
| 9% positive |
| 91% negative |
|
|
|
|
|
|
|
|
| p(breast cancer | test positive) = (0.01 * 0.9) / (0.01 * 0.09) + (0.99 * 0.09) | ||||||
o Natural frequency
|
|
|
| 1000 women |
|
|
|
|
| 10 breast cancer |
|
|
| 990 no breast cancer |
|
| 9 positive |
| 1 negative |
| 89 positive |
| 901negative |
|
|
|
|
|
|
|
|
| p(breast cancer | test positive) = 9 /( 9 + 89) | ||||||
- Many clinicians do not know probability person has disease given a positive screening test.
o positive predictive value
Higher Survival Does Not Mean Longer Life
- meaningless in making comparisons across groups of people that differ dramatically in how diagnosis is made.
- 5 year survival rate
o number of patients diagnosed with disease still alive after 5 years / number of patients diagnosed.
- annual mortality rate
o number of people who die of disease in year / number of people in group
- diagnosed appears in numerator and denominator of 5 year survival rate, but no where in mortality rate.
- screening profoundly biases survival
o affects timing of diagnosis (lead time bias)
o affects nature of diagnosis by including people with non progressive cancers (over diagnosis bias)
- survival rates can be increased by setting the time of diagnosis earlier, even if no life is prolonged / saved.
What is Statistical Literacy?
- Learning to live with uncertainty – understand there is no certainty and no zero risk, but only risks that are more or less acceptable
- Questions to ask of risk
o risk of what
o time frame?
o how big
o does it apply to me
o screening tests
§ understand screening tests may have benefits and harms
§ understand screening tests can make two errors
· false positives
· false negatives
§ understand how to translate specificities, sensitivities and other conditional probabilities into natural frequencies
§ understand goal of screening is not simply early detection of disease, it is mortality reduction or improvement of life
o treatments
§ understand that treatments typically have benefits and harms
§ understand size of benefit and harm
o Questions about science behind numbers
§ quality of evidence
§ conflicts of interest
- screening is itended to detect existing cancers at early stage à so it does not reduce risk of getting breast cancern
- patients were asked what do you feel is the likelihood of you having a heart attack over next 12 months
o likelihood depends on individual factors like
§ age / sex/ smoking / diabetes
o patients risk estimates showed no correlation with these factors
- relative risk reductions can cause exaggerated perceptions of treatment effects
- do clinicians understand the number needed to treat, which is
o number of patients that must be treated in order to save the life of one patient.
- 4 identical funding proposals à cardiac rehab and breast cancer screening
o relative risk reduction
o absolute risk reduction
o absolute values from which the absolute risk reduction is computed
o number needed to treat
§ relative risk reduction – seen as having greatest merit
- geography is destiny
o surgical treatments are often not based evidence
- speciality is destiny
- informed consent refers to ideal how doctors and patients interact.
- asubtle way to induce the illusion of certainity – by analogies
o war on cancer
- prevalent use of relative risk
o sometimes defended on basis that ratio measures are transportable to different populations with different baseline risks
§ this is main weakness as well, since they conceal underlying absolute risks
- some citizens believe à 30% probability of rain tomorrow
o will rain tomorrow 30% of time
o in 30% of the area
o on 30% of days for which announcement was made
è problem – no reference class is used
Prozac example
psychaiatrist – thinking of all his patients
patients thought of themselves alone
sensitivities / specificities à two reference classes
patients with disease
patients without disease
natural frequencies à same reference class
è all patients
No comments:
Post a Comment