Transcript The Medical Exam History - Lifelong Learning Academy

Summary
1. Evidence is important
2. Evidence is hard to come by
3. The law of diminishing returns applies to
prevention
4. RCT (Randomized controlled trial) is the
gold standard
5. RRR (relative risk reduction)
6. ARR (absolute risk reduction)
7. NNT (number needed to treat)
Dubious Attributions 1
•
•
•
•
•
•
•
•
The data suggests…
Maybe
Associated with
Preliminary or “pilot study”
Not peer reviewed
Retrospective
Generally accepted
Tradition
Dubious Attributions 2
•Not randomized
•Not controlled ( ignores natural course of disease)
•This study funded by…
•Cannot exclude (the absence of proof is not proof of
absence)
•Funding ( authors are paid consultants)
•Data based on telephone or questionnaire interview
•Zeal – it is well documented that enthusiasm is
inversely proportional to the quality of the data
Relative vs. Absolute
• Relative risk reduction –
compares the % change of one
Rx with another. E.g. treatment
A decreases mortality by 1% and
treatment B by 2%: RRR 50%
• Absolute risk reduction – 2%-1%
equals 1% (may not be clinically
relevant) .
NNT
• Take the difference in the rate of response,
death or any other criterion in the control
group minus the rate in the treated or
tested group.
• E.g. 8% in control group and 4% in treated
group. The difference is 4%. So for every
100 patients (subjects) four will be
benefited (4%) i.e. NNT = 25 to benefit
one patient.
science
• A mode of inquiry, a method of asking
appropriate questions that themselves are
falsifiable.
• Data must be independently reproducible
• Conclusions are derived from the data produced
by the method and are subject to statistical
analysis and sometimes subjective
interpretation.
• The questions are “how and/or what” not “why”.
The “why” question is left to metaphysics,
philosophy and religion.
Medical model
Symptoms
Signs
Tests
Diagnosis
Prognosis (natural history e.g. The
Tuskegee Study)
Treatment
Medical exam
• History : Chief complaint, present illness,
past history, social history, family history
• Physical Exam
• Tests (lab, imaging, invasive)
• Make Diagnosis
• Prognosis ( natural history of illness)
• Treatment
HOW TO CHOOSE A DR.
• KNOWLEDGE (Board certification and
CME)
• EMPATHY (listens well, answers your
questions)
• FIRST YOUR WELL-BEING (not boat or
car payment). Cares for you.
• 2/3 are not science related
Questions to ask
• Is the treatment routine for the specific
disease or condition?
• What are the benefits to be expected, how
will they be measured, and how long will it
take to see results?
• What is the scientific evidence that the
treatment will work and where was it
published?
• What are the risks or adverse effects?
More questions to ask
• Why am I getting this drug?
• What are the risks vs. the
benefits?
• Is there an older (tried and true)
drug or lifestyle change that
works just as well?
• Will it interfere with my other
medications?
Biological systems
• Never say never
• Never say always
• Don’t be the first nor the last to use a
new treatment
• Specialist: Knows more and more
about less and less until he knows
everything about nothing
• Generalist: knows less and less
about more and more until he knows
nothing about everything
Questions of Research
• Relevance – does the research
allow us to live longer, happier or
healthier.
• Validity – Randomized, double
blind, controlled trial or study
• Clinically important – does it
apply to you, ARR, NNT.
Questions of the Dr.
• What is the natural history of my
condition? (what happens if I do nothing)
• What is the probability of this course of
action being beneficial and/or harmful to
me?
• What is the time course of treatment
(acute or chronic)?
• What is the cost?
Post hoc ergo propter hoc
(after the fact therefore the
cause of the fact, a fallacious
argument)
Primum non nocere
There is no medical problem,
however bad, that cannot be made
worse by surgical intervention