Session 6: Evidence-Based Medicine Flashcards
How does biological plausibility relate to causal link?
Causality is more likely if a biologically plausible mechanism is likely or demonstrated.
This means that a biological plausible mechanism strengthens the case for a casual link.
Give examples where biological plausibility strengthening causality is limited.
It is limited by the current knowledge. An example is citrus fruit was shown to prevent scurvy before vitamin C was discovered.
It also does not guarantee an association or effect.
Define evidence-based medicine.
The conscientious, explicit and judicious use of current best evidence in making decisions about the care of individual patients.
This means integrating individual clinical expertise with the best available external clinical evidence.
Explain limitations to evidence-based medicine.
It is not meant to be a ‘cookbook’ for medicine.
External clinical evidence can inform but can never replace individual clinical expertise.
The individual clinical expertise is important to decide whether the external evidence is appropriate and applies to the patient.
Triad of evidence-based medicine.
A mix of the clinician’s clinical judgment, the relevant scientific evidence and the patient’s values and preferences.
How can scientifically proven to work be interpreted?
You are more likely to benefit than not.
It is better than the other treatment.
Limitations of ‘you are more likely to benefit than not’.
It assumes that the majority of people receiving this treatment are cured.
However that is not necessarily the case, and it can still work.
But it might still be better than another treatment.
A study has 755 people receiving a CVS drug and 745 receiving a placebo.
63 people receiving drug dies.
26 people receiving placebo dies.
What is the relative risk?
63/755=8.34%
26/743=3.50%
8.34%/3.50%=2.38
The relative risk is 2.38
What does a relative risk of 2.38 mean?
That you are 2.38 times more likely to die if you take the CVS drug than if on placebo.
Limitations of relative risk.
It only tells you are more likely e.g. in 2.38.
But the likelihood might be very small (1% -> 2%) or may be very high (30% -> 60%).
So a high relative risk might not mean that you are much more likely if the risk is very small to begin with.
In order to know a bit more about the study and not just rely on relative risk, what is calculated?
Risk difference.
What is the risk difference of the study?
(8.34% and 3.50%)
8.34% - 3.50% = 4.84%.
What is the absolute risk of the people receiving the drugs where 63 out of 755 died?
63/755 = 8.34%
This means that you are 8.34% likely to die when on the drug.
Calculate the number needed to harm (NNH).
100/4.84 (risk difference)
= 21
What does the NNH tell you? (21)
It is a comparisonof the two arms of the trial and not the same as absolute risk.
This is a transfer from drug to placebo.
It means how many are needed to be treated with the CVS drug compared to the placebo in order to have an adverse effect.
