Stats Flashcards
Statistical hypotheses, 2 types
null=Hypothesis of no difference or relationship
Alternative (H1) Hypothesis of some difference or relationship
Type I error (α)
α = you “a”bserved a difference that did not exist
This is also the a we compare p to
p< a= reject the null
p> a= accept the null
Type II error (β)
β = you were blinded by the truth.
Power = 1-B
If you ↑ sample size, you ↑ power
If p < 0.05, then there is ____ chance that the data will show something that is not really there.
less than a 5%
Number needed to treat
NNT= 1/ARR
ARR= absolute risk reduction (difference in risk)
Respondent bias
outcome is obtained by pt response, like migraines
outcome cannot be objectively measured
Hazard ratio
close to 1= the hazard between groups is about the same
test group HR > 1 is higher chance
test group HR < 1 is a lower chance
Achilles tendon in old people
reflex may be decreased/ absent
Labs in acute mesenteric ischemia
Metabolic acidosis
↑ WBC
↑ hemoglobin
↑ amylase
Lymphoblastic leukemia
↑↑↑ blasts on smear
acutely ill from other ↓ cell lines (bleeding, infection, B s/s)
2 options as 1st line agents in bipolar
Li
valproate if renal dysfxn
Most common symptoms of sickle cell trait (< 50% Hs) is
Painless hematuria
other sickle cell disease s/s are unlikely
Cell free fetal DNA
pregos > 35 yo
1st trimester
very sensitive/specific test
Cyclophosphamide on cancer man
bladder- carcinoma and hemorrhagic cystitis
HCM murmur intensity
↓ preload–> ↑ intensity
standing
